Istituto Universitario di Studi Superiori

Universit degli Studi di Pavia

ROSE SCHOOL European School for Advanced Studies in Reduction of Seismic Risk

Evaluation of pushover procedures for the seismic design of buildings

A Thesis Submitted in Partial Fulfilment of the Requirements for the Master Degree in EARTHQUAKE ENGINEERING
by

Dario Pietra Supervisor: Dr. Rui Pinho

2008

The dissertation entitled Evaluation of pushover procedures for the seismic design of buildings, by Dario Pietra, has been approved in partial fulfilment of the requirements for the Master Degree in Earthquake Engineering.

Dr. Rui Pinho_____________________

ABSTRACT

A number of recent studies raised doubts on the effectiveness of conventional pushover methods, whereby a constant incremental force vector is applied to the structure, in estimating the seismic demand/capacity of framed buildings subjected to earthquake action. The latter motivated the recent development of the so-called Adaptive Pushover methods whereby the loading vector is updated at each analysis step, considering one or more response modes, reflecting in this way the effects that damage progression have on the response characteristics of structures subjected to increasing loading levels. Within such adaptive framework, the application of a displacement incremental loading vector becomes not only feasible but also possibly advantageous since it seems to lead to superior response predictions, with little or no additional modelling/analysis effort, with respect to conventional pushover procedures. In this work, a parametric study, whereby the accuracy of the Displacement-based Adaptive
Pushover algorithm (DAP) in predicting the seismic response of 3-, 9- and 20-storey high steel buildings responding in the inelastic range is presented. A large set of natural records is used in the dynamic analyses that are carried out for comparison. The performance of the adaptive procedures is evaluated in terms of prediction of the main structural response parameters of interest (interstorey drifts, shears and overturning moments). Results, expressed as absolute values of these design parameters as well as their ratio between static- and dynamic-analysis values, are compared with those provided by conventional pushover schemes. Results show that DAP, compared with non-adaptive procedures, represents an alternative simpler procedure (involving a single pushover analysis) that allows predicting the response shape of high-rise steel buildings with an accuracy that is at least as good as that obtained with more complex multiple-pushover procedures.

ii

iii

ACKNOWLEDGMENTS

The author wishes to thank in particular all those people whose friendly assistance and wise guidance supported him throughout the duration of this research project. Thanks to the Professors and Technicians of the Department of Structural Mechanics at the University of Pavia and the Professors of the Rose School. Thanks particularly to Dr. R. Pinho, for proposing me such an interesting research topic and for being always willing and available to help. The author would like to acknowledge the invaluable assistance of Dr. Mark Aschheim, who kindly supplied data on the structural models and input motion employed and described in the FEMA-440 report. Finally, the technical support from Dr. Stelios Antoniou is gratefully acknowledged.

iv

Table of Contents
ABSTRACT........................................................................................................................................................ i ACKNOWLEDGMENTS............................................................................................................................iii Table of Contents .............................................................................................................................................v List of Figures ..................................................................................................................................................ix List of Tables...................................................................................................................................................xv 1 INTRODUCTION .....................................................................................................................................1 1.1 Forward ................................................................................................................................................1 1.2 Aims of the Work ...............................................................................................................................3 1.3 Thesis Outline......................................................................................................................................4 2 Nonlinear Static Procedures: Critical Overview ......................................................................................5 2.1 Introduction .........................................................................................................................................5 2.2 NSPs in Earthquake Engineering .....................................................................................................7 2.2.1 The Eurocode 8 (prENV 1998-1, 1994) ................................................................................8 2.2.2 The Coefficient Method in FEMA 356 (ASCE, 2000) ........................................................9 2.2.3 The Capacity Spectrum Method in ATC 40 (ATC, 1996) .................................................10 2.3 Pushover Analysis in Earthquake Engineering.............................................................................11 2.3.1 Non-Adaptive Non-Modal Procedures (NANM Procedures) .........................................13 2.3.2 Non-Adaptive Modal Procedures (NAM Procedures).......................................................14 2.3.3 Adaptive Procedures................................................................................................................17 2.3.3.1 Single-run adaptive pushover analyses .............................................................................21 2.3.3.2 Multiple-run adaptive pushover analyses .........................................................................22 2.3.4 Discussion .................................................................................................................................24 2.3.4.1 Selection of the control node.............................................................................................25 2.3.4.2 Alternative modal combination rules ...............................................................................25 2.3.4.3 Drawbacks in multiple-run analyses .................................................................................27 2.3.4.4 Multiple Degrees Of Freedom Effects (MDOFEs).......................................................27 2.3.4.5 Innovative strategies............................................................................................................32 2.3.5 Adaptive Pushover Algorithm ...............................................................................................34 3 Case Studies and Analyses Post-Processing ...........................................................................................41 3.1 Introduction .......................................................................................................................................41 3.2 Case studies ........................................................................................................................................41 3.2.1 First parametric study..............................................................................................................41 3.2.1.1 3-Storey Frame.....................................................................................................................43 3.2.1.2 9-Storey Frame.....................................................................................................................44 3.2.2 Second parametric study .........................................................................................................45 3.2.2.1 20-Storey frame....................................................................................................................46 3.3 Analytical tool ....................................................................................................................................49 3.3.1 Modeling of frames..................................................................................................................49 3.3.2 Verification of the structural analysis software for steel frames .......................................50 3.3.2.1 Case study .............................................................................................................................50 3.3.2.2 The model.............................................................................................................................51 3.3.2.3 Results ...................................................................................................................................53 3.3.2.4 Conclusions ..........................................................................................................................54 3.4 Ground motions................................................................................................................................55 3.4.1 First parametric study..............................................................................................................55 3.4.2 Second parametric study .........................................................................................................58 3.5 The Analyses......................................................................................................................................59 3.5.1 First parametric study..............................................................................................................59

vi

3.5.2 Second parametric study......................................................................................................... 61 3.6 Response Statistics............................................................................................................................ 62 3.6.1 First parametric study.............................................................................................................. 62 3.6.2 Second parametric study......................................................................................................... 64 4 Capacity Curves.......................................................................................................................................... 65 4.1 Introduction....................................................................................................................................... 65 4.2 Ordinary Ground Motions .............................................................................................................. 65 4.2.1 IDA curves ............................................................................................................................... 66 4.2.2 Qualitative evaluation of the accuracy of the static procedures........................................ 67 4.2.3 Quantitative evaluation of the accuracy of the static procedures ..................................... 69 4.3 Near-Field records ............................................................................................................................ 71 4.3.1 IDA curves ............................................................................................................................... 71 4.3.2 Qualitative evaluation of the accuracy of the static procedures........................................ 72 4.3.3 Quantitative evaluation of the accuracy of the static procedures ..................................... 73 4.4 Conclusions........................................................................................................................................ 75 5 Storey Response Parameters .................................................................................................................... 77 5.1 Introduction....................................................................................................................................... 77 5.2 First parametric study....................................................................................................................... 77 5.2.1 Overview of FEMA 440 results ............................................................................................ 78 5.2.1.1 Ordinary Ground Motions ................................................................................................ 78 5.2.1.2 Near-Field Motions.............................................................................................................79 5.2.1.3 Conclusions.......................................................................................................................... 79 5.2.2 Evaluation of results................................................................................................................ 80 5.2.3 Ordinary Ground Motions ..................................................................................................... 80 5.2.3.1 Storey Displacements ......................................................................................................... 82 5.2.3.2 Interstorey Drifts................................................................................................................. 83 5.2.3.3 Interstorey Shears and Moments ...................................................................................... 85 5.2.4 Near-Field Motions ................................................................................................................. 87 5.2.4.1 Storey Displacements ......................................................................................................... 87 5.2.4.2 Interstorey Drifts................................................................................................................. 88 5.2.4.3 Inter- storey Shears and Moments.................................................................................... 91 5.2.5 Conclusions .............................................................................................................................. 92 5.3 Second parametric study .................................................................................................................. 94 5.3.1 Overview of MPA results ....................................................................................................... 94 5.3.2 Evaluation of results................................................................................................................ 96 5.3.3 Preliminary study ..................................................................................................................... 96 5.3.4 Extensive study ........................................................................................................................ 99 5.3.4.1 Displacement correction factor method.......................................................................... 99 5.3.4.2 Scaling factor method.......................................................................................................102 5.3.5 Conclusions ............................................................................................................................105 6 Conclusions...............................................................................................................................................107 6.1. Summary...........................................................................................................................................107 6.2. Future Research ..............................................................................................................................111 References ..........................................................................................................................................................113 Appendix A: Prototype Buildings...................................................................................................................119 (members section and frame dynamic properties) .......................................................................................119 A.1 Section properties ...........................................................................................................................119 A.2 Dynamic properties ........................................................................................................................121 Appendix B: First parametric study scaling factors ..................................................................................123 Appendix C: First parametric study - Capacity Curves ...............................................................................125 C.1 OGMs...............................................................................................................................................125

vii

C.1.1 Representations of IDA envelopes .....................................................................................125 C.1.2 Pushover curves .....................................................................................................................127 C.1.3 Mean and maximum values of E1 .......................................................................................128 C.1.4 E2 .............................................................................................................................................129 C.2 NF records .......................................................................................................................................130 C.2.1 3-Storey frame ........................................................................................................................130 C.2.2 3-Storey weak frame ..............................................................................................................133 C.2.3 9-Storey frame ........................................................................................................................136 C.2.4 9-Storey weak frame ..............................................................................................................139 Appendix D: First parametric study - Storey response parameters ...........................................................143 D.1 OGMs ...............................................................................................................................................143 D.1.1 3-Storey frame ........................................................................................................................143 Fig. 104. 3-Storey frame-total drift level 0.5%.....................................................................................144 D.1.2 3-Storey weak frame ..............................................................................................................147 D.1.3 9-Storey frame ........................................................................................................................150 D.1.4 9-Storey weak frame ..............................................................................................................153 D.2 NF records .......................................................................................................................................156 D.2.1 3-Storey frame ........................................................................................................................157 D.2.2 3-Storey weak frame ..............................................................................................................161 D.2.3 9-Storey frame ........................................................................................................................165 D.2.4 9-Storey weak frame ..............................................................................................................169 Appendix E: First parametric study - error measurement (OGMs) ..........................................................173 E.1 Mean and Maximum E1.................................................................................................................173 E.1.1 3-Storey frame ........................................................................................................................173 E.1.2 3-Storey weak frame ..............................................................................................................175 E.1.3 9-Storey frame ........................................................................................................................178 E.1.4 9-Storey weak frame ..............................................................................................................180 E.2 Values of E1 and E2 for each building model............................................................................182 E.2.1 3-Storey frame ........................................................................................................................182 E.2.2 3-Storey weak frame ..............................................................................................................186 E.2.3 9-Storey frame ........................................................................................................................189 E.2.4 9-Storey weak frame ..............................................................................................................192 Appendix F: First parametric study - error measurement (NF records)...................................................195 F.1 Mean and maximum E1.................................................................................................................195 F.2 3-Storey frame .................................................................................................................................196 F.3 3-Storey weak frame .......................................................................................................................202 F.4 9-Storey frame .................................................................................................................................207 F.5 9-Storey weak frame .......................................................................................................................212 Appendix G: Second parametric study ..........................................................................................................219 G.1 Preliminary study.............................................................................................................................219 G.1.1 Storey response parameters ..................................................................................................219 G.1.2 Bias ...........................................................................................................................................221 G.1.3 Dispersion ...............................................................................................................................223 G.2 Extensive study Displacement correction factor method ....................................................225 G.2.1 Storey response parameters ..................................................................................................225 G.2.2 Bias ...........................................................................................................................................227 G.2.3 Dispersion ...............................................................................................................................229 G.3 Extensive study Scaling factor method ...................................................................................231 G.3.1 Storey response parameters ..................................................................................................231 G.3.2 Bias ...........................................................................................................................................233

viii

ix

List of Figures
Fig. 1. Equivalent Single Degree Of Freedom System ......................................................................................9 Fig. 2. Coefficient Method (ASCE, 2000) ........................................................................................................ 10 Fig. 3. Capacity Spectrum Method (ATC, 1996) ............................................................................................. 11 Fig. 4. Properties of the nth mode inelastic SDOF system derived from the corresponding pushover curve.............................................................................................................................................................. 16 Fig. 5. Periods of vibration of a 4-Storey structure under increasing levels of deformation .................... 18 Fig. 6. Interstorey drift profile of a 12-Storey building subjected to increasing levels of deformation... 19 Fig. 7. Adaptive pushover: shape of loading vector is updated at each analysis step................................. 19 Fig. 8. Bilinear idealization of a modal capacity diagram at the ith pushover step ...................................... 24 Fig. 9. Storey force distributions of a 12-Storey building obtained with Displacement-based Adaptive Pushover as well with standard non-adaptive pushovers...................................................................... 34 Fig. 10. Correlation Coefficient-Frequency Ratio relationship...................................................................... 38 Fig. 11. Updating of the loading displacement vector .................................................................................... 39 Fig. 12. Incremental Updating strategy ............................................................................................................. 40 Fig. 13. 3-Storey (left) and 9-Storey (right) steel frame structures: plane view ........................................... 43 Fig. 14. 3-Storey steel frame: vertical view ....................................................................................................... 43 Fig. 15. 3 Storey steel frame-1st mode pushover SeismoStruct (left), FEMA 440 (right) .......................... 44 Fig. 16. 9-Storey steel frame-1st mode pushover SeismoStruct (left), FEMA 440 (right).......................... 44 Fig. 17. 9-Storey steel frame: vertical view ....................................................................................................... 45 Fig. 18. 20-Storey steel frame- SAC 1st mode (Model M1) and SeismoStruct 1st mode (red) pushover curves ............................................................................................................................................................ 46 Fig. 19. 20-Storey steel frame: plan view and elevation (from Ohtori et al., 2003)...................................................... 47 Fig. 20. 3-Storey (left) and 9-Storey (right) frame: mode shapes................................................................... 48 Fig. 21. 20-Storey frame: mode shapes ............................................................................................................. 48 Fig. 22. Quasi-static loading program................................................................................................................ 51 Fig. 23. Elevation of the test structure (unit: mm) ...................................................................................................... 51 Fig. 24. Plan of the test structure (unit: mm) ................................................................................................... 52 Fig. 25. Through diaphragm connection (left) and detail of the column base (right) (unit: mm)............ 52 Fig. 26. Total base shear vs. total drift ...................................................................................................................... 54 Fig. 27. Storey shear vs. storey drift 1st storey .............................................................................................. 55 Fig. 28. Storey shear vs. storey drift 2nd storey ............................................................................................. 55 Fig. 29. Ordinary Ground Motions: unscaled acceleration response spectrum.......................................... 56 Fig. 30. Ordinary Ground Motions: unscaled displacement response spectrum........................................ 56 Fig. 31. Maximum base shear and top floor displacement values obtained with incremental dynamic analyses ......................................................................................................................................................... 59 Fig. 32. Alternative representation of IDA results .......................................................................................... 67 Fig. 33. 3-Storey frame ....................................................................................................................................... 68 Fig. 34. 9-Storey weak frame .............................................................................................................................. 68 Fig. 35. 9-Storey frame ....................................................................................................................................... 69 Fig. 36. 3-Storey frame ....................................................................................................................................... 71 Fig. 37. 9-Storey frame ....................................................................................................................................... 71 Fig. 38. 3-Storey frame ........................................................................................................................................ 71 Fig. 39. 3-Storey frame ........................................................................................................................................ 72 Fig. 40. 9-Storey frame ........................................................................................................................................ 73 Fig. 41. 9-Storey frame ........................................................................................................................................ 73 Fig. 42. 3-Storey frame (SCHMV1) .................................................................................................................. 74 Fig. 43. 9-Storey frame (LUCMV1)................................................................................................................... 74 Fig. 44. 9-Storey weak frame (LUCMV1)......................................................................................................... 74

Fig. 45. 9-Storey weak frame (LUCMV1)..........................................................................................................74 Fig. 46. Coefficient of variation of dynamic analyses results: (a) floor displacements, (b) interstorey drifts, (c) interstorey shears, (d) interstorey moments ............................................................................81 Fig. 47. Coefficient of variation of DAP analyses results: (a) floor displacements, (b) interstorey drifts, (c) interstorey shears, (d) interstorey moments .......................................................................................81 Fig. 48. 9-Storey weak frame ...............................................................................................................................83 Fig. 49. 9-Storey frame .........................................................................................................................................85 Fig. 50. 9-Storey frame-drift level 4%: Interstorey shear (E2) .......................................................................86 Fig. 51. 9-Storey frame-RRSMV1.......................................................................................................................89 Fig. 52. 9-Storey frame-SCHMV1 ......................................................................................................................90 Fig. 53. 9-Storey frame-SCHMV1: Interstorey drift (E2) ...............................................................................91 Fig. 54. 9-Storey frame-RRSMV1: Interstorey shear (E2) ..............................................................................92 Fig. 55. Mean and standard deviation along the height of the buildings for the whole set of models at the intermediate deformation level (E1)...................................................................................................93 Fig. 56. Mean values among the whole set of models for all the NF records (E1).....................................94 Fig. 57. Interstorey drift profile (from Goel and Chopra, 2004) ...................................................................95 Fig. 58. Interstorey drift ratio (from Goel and Chopra, 2004).......................................................................95 Fig. 59. Interstorey drift: bias for the 9-Storey (left) and 20-Storey (right) frames .....................................98 Fig. 60. Interstorey overturning moment: bias for the 9-Storey (left) and 20-Storey (right) frames ........98 Fig. 61. Bias in interstorey drift: median values along the height of the structure for the whole set of structural models adopted in both the first and the second parametric studies and intermediate value adopted (red line).............................................................................................................................100 Fig. 62. Interstorey drift profile: 9-Storey (left) and 20-Storey (right) frames............................................101 Fig. 63. Interstorey drift: bias for the 9-Storey (left) and 20-Storey (right) frames ...................................101 Fig. 64. Interstorey overturning moment profile: 9-Storey (left) and 20-Storey (right) frames...............102 Fig. 65. Corrected interstorey drift profile: 9-Storey frame (left) and 20-Storey frame (right)............103 Fig. 66. Corrected interstorey drift ratio (bias): 9-Storey frame (left) and 20-Storey frame (right).....104 Fig. 67. Corrected bias in the interstorey shear for the 9-Storey frame (left) and bias in the interstorey moment for the 20-Storey frame (right) .................................................................................................104 Fig. 68. Corrected interstorey moment profile for the 9-Storey frame (left) and interstorey shear profile for the 20-Storey frame (right) ....................................................................................................105 Fig. 69. Steel cross section parameters.............................................................................................................120 Fig. 70. 3-Storey frame .......................................................................................................................................125 Fig. 71. 3-Storey weak frame .............................................................................................................................126 Fig. 72. 9-Storey frame .......................................................................................................................................126 Fig. 73. 9-Storey weak frame .............................................................................................................................126 Fig. 74. 3-Storey frame .......................................................................................................................................127 Fig. 75. 3-Storey weak frame .............................................................................................................................127 Fig. 76. 9-Storey frame .......................................................................................................................................127 Fig. 77. 9-Storey weak frame .............................................................................................................................128 Fig. 78. E1-Mean and Maximum values ..........................................................................................................128 Fig. 79. E2 ............................................................................................................................................................129 Fig. 80. 3-Storey frame-NF records-Capacity Curves ERZMV1 .............................................................130 Fig. 81. 3-Storey frame-NF records-Capacity Curves LUCMV1 .............................................................130 Fig. 82. 3-Storey frame-NF records-Capacity Curves RRSMV1 ..............................................................131 Fig. 83. 3-Storey frame-NF records-Capacity Curves SCHMV1..............................................................131 Fig. 84. 3-Storey frame-NF records-E1 ...........................................................................................................132 Fig. 85. 3-Storey frame-NF records-E2 ...........................................................................................................132 Fig. 86. 3-Storey weak frame-NF records-Capacity Curves ERZMV1 ...................................................133 Fig. 87. 3-Storey weak frame-NF records-Capacity Curves LUCMV1 ....................................................133 Fig. 88. 3-Storey weak frame-NF records-Capacity Curves RRSMV1 ....................................................134

xi

Fig. 89. 3-Storey weak frame-NF records-Capacity Curves SCHMV1................................................... 134 Fig. 90. 3-Storey-weak frame-NF records-E1................................................................................................ 135 Fig. 91. 3-Storey-weak frame-NF records-E2................................................................................................ 135 Fig. 92. 9-Storey frame-NF records-Capacity Curves ERZMV1............................................................. 136 Fig. 93. 9-Storey frame-NF records-Capacity Curves LUCMV11 .......................................................... 136 Fig. 94. 9-Storey frame-NF records-Capacity Curves RRSMV1 ............................................................. 137 Fig. 95. 9-Storey frame-NF records-Capacity Curves SCHMV1............................................................. 137 Fig. 96. 9-Storey frame-NF records-E1 .......................................................................................................... 138 Fig. 97. 9-Storey frame-NF records-E2 .......................................................................................................... 138 Fig. 98. 9-Storey weak frame-NF records-Capacity Curves ERZMV1 .................................................. 139 Fig. 99. 9-Storey weak frame-NF records-Capacity Curves LUCMV1................................................... 139 Fig. 100. 9-Storey weak frame-NF records-Capacity Curves RRSMV1 ................................................. 140 Fig. 101. 9-Storey weak frame-NF records-Capacity Curves SCHMV1................................................. 140 Fig. 102. 9-Storey-weak frame-NF records-E1.............................................................................................. 141 Fig. 103. 9-Storey-weak frame-NF records-E2.............................................................................................. 141 Fig. 104. 3-Storey frame-total drift level 0.5% ............................................................................................... 144 Fig. 105. 3-Storey frame-total drift level 2% .................................................................................................. 145 Fig. 106. 3-Storey frame-total drift level 4% .................................................................................................. 146 Fig. 107. 3-Storey-weak frame-total drift level 0.5%..................................................................................... 147 Fig. 108. 3-Storey-weak frame-total drift level 2%........................................................................................ 148 Fig. 109. 3-Storey-weak frame-total drift level 4%........................................................................................ 149 Fig. 110. 9-Storey frame-total drift level 0.5% ............................................................................................... 150 Fig. 111. 9-Storey frame-total drift level 2% .................................................................................................. 151 Fig. 112. 9-Storey frame-total drift level 4% .................................................................................................. 152 Fig. 113. 9-Storey weak frame-total drift level 0.5% ..................................................................................... 153 Fig. 114. 9-Storey weak frame-total drift level 2% ........................................................................................ 154 Fig. 115. 9-Storey weak frame-total drift level 2.7% ..................................................................................... 155 Fig. 116. 3 Storey frame-ERZMV1 ................................................................................................................. 157 Fig. 117. 3 Storey frame-LUCMV1 ................................................................................................................. 158 Fig. 118. 3 Storey frame-RRSMV1 .................................................................................................................. 159 Fig. 119. 3 Storey frame-SCHMV1.................................................................................................................. 160 Fig. 120. 3-Storey weak-frame-ERZMV1....................................................................................................... 161 Fig. 121. 3-Storey-weak frame-LUCMV1....................................................................................................... 162 Fig. 122. 3-Storey-weak frame-RRSMV1........................................................................................................ 163 Fig. 123. 3-Storey-weak frame-SCHMV1 ....................................................................................................... 164 Fig. 124. 9-Storey -frame-ERZMV1................................................................................................................ 165 Fig. 125. 9-Storey -frame-LUCMV1................................................................................................................ 166 Fig. 126. 9-Storey -frame-RRSMV1................................................................................................................. 167 Fig. 127. 9-Storey -frame-SCHMV1................................................................................................................ 168 Fig. 128. 9-Storey weak-frame-ERZMV1....................................................................................................... 169 Fig. 129. 9-Storey weak-frame-LUCMV1....................................................................................................... 170 Fig. 130. 9-Storey weak-frame-RRSMV1........................................................................................................ 171 Fig. 131. 9-Storey weak-frame-SCHMV1 ....................................................................................................... 172 Fig. 132. 3-Storey frame-E1 (maximum values) ............................................................................................ 173 Fig. 133. 3-Storey frame-E1 (mean values)..................................................................................................... 174 Fig. 134. 3-Storey weak frame-E1 (maximum values) .................................................................................. 176 Fig. 135. 3-Storey weak frame-E1 (mean values) .......................................................................................... 176 Fig. 136. 9-Storey frame-E1 (maximum values) ............................................................................................ 178 Fig. 137. 9-Storey frame-E1 (mean values)..................................................................................................... 179 Fig. 138. 9-Storey weak frame-E1 (maximum values) .................................................................................. 180 Fig. 139. 9-Storey weak frame-E1 (mean values) ......................................................................................... 181

xii

Fig. 140. 3-Storey frame-total drift level 0.5%-E1 .........................................................................................183 Fig. 141. 3-Storey frame-total drift level 0.5%-E2 .........................................................................................183 Fig. 142. 3-Storey frame-total drift level 2%-E1 ............................................................................................184 Fig. 143. 3-Storey frame-total drift level 2%-E2 ............................................................................................184 Fig. 144. 3-Storey frame-total drift level 4%-E1 ............................................................................................185 Fig. 145. 3-Storey frame-total drift level 4%-E2 ............................................................................................185 Fig. 146. 3-Storey weak-frame-total drift level 0.5%-E1...............................................................................186 Fig. 147. 3-Storey weak-frame-total drift level 0.5%-E2...............................................................................186 Fig. 148. 3-Storey weak-frame-total drift level 2%-E1 ..................................................................................187 Fig. 149. 3-Storey weak-frame-total drift level 2 %-E2.................................................................................187 Fig. 150. 3-Storey weak-frame-total drift level 4%-E1 ..................................................................................188 Fig. 151. 3-Storey weak-frame-total drift level 4%-E2 ..................................................................................188 Fig. 152. 9-Storey frame-total drift level 0.5%-E1 .........................................................................................189 Fig. 153. 9-Storey frame-total drift level 0.5%-E2 .........................................................................................189 Fig. 154. 9-Storey frame-total drift level 2%-E1 ............................................................................................190 Fig. 155. 9-Storey frame-total drift level 2%-E2 ............................................................................................190 Fig. 156. 9-Storey frame-total drift level 4%-E1 ............................................................................................191 Fig. 157. 9-Storey frame-total drift level 4%-E2 ............................................................................................191 Fig. 158. 9-Storey weak-frame-total drift level 0.5%-E1...............................................................................192 Fig. 159. 9-Storey weak-frame-total drift level 0.5%-E2...............................................................................192 Fig. 160. 9-Storey weak-frame-total drift level 2%-E1 ..................................................................................193 Fig. 161. 9-Storey weak-frame-total drift level 2%-E2 ..................................................................................193 Fig. 162. 9-Storey weak-frame-total drift level 2.7%-E1...............................................................................194 Fig. 163. 9-Storey weak-frame-total drift level 2.7%-E2...............................................................................194 Fig. 164. 3-Storey frame .....................................................................................................................................195 Fig. 165. 3-Storey weak frame ...........................................................................................................................195 Fig. 166. 9-Storey frame .....................................................................................................................................196 Fig. 167. 9-Storey weak frame ...........................................................................................................................196 Fig. 168. 3-Storey frame-E1-Floor Displacement ..........................................................................................197 Fig. 169. 3-Storey frame-E1-Interstorey Drift................................................................................................197 Fig. 170. 3-Storey frame-E1-Interstorey Shear ...............................................................................................198 Fig. 171. 3-Storey frame-E1-Interstorey Moment..........................................................................................199 Fig. 172. 3-Storey frame-E2-Floor Displacement ..........................................................................................199 Fig. 173. 3-Storey frame-E2-Interstorey Drift................................................................................................200 Fig. 174. 3-Storey frame-E1-Interstorey Shear ...............................................................................................201 Fig. 175. 3-Storey frame-E1-Interstorey Moment..........................................................................................201 Fig. 176. 3-Storey-weak frame-E1-Floor Displacement................................................................................202 Fig. 177. 3-Storey-weak frame-E1-Interstorey Drift......................................................................................203 Fig. 178. 3-Storey-weak frame-E1-Interstorey Shear.....................................................................................203 Fig. 179. 3-Storey-weak frame-E1-Interstorey Moment ...............................................................................204 Fig. 180. 3-Storey-weak frame-E2-Floor Displacement................................................................................205 Fig. 181. 3-Storey-weak frame-E2-Interstorey Drift......................................................................................205 Fig. 182. 3-Storey-weak frame-E1-Interstorey Shear.....................................................................................206 Fig. 183. 3-Storey-weak frame-E1-Interstorey Moment ...............................................................................207 Fig. 184. 9-Storey frame-E1-Floor Displacement ..........................................................................................207 Fig. 185. 9-Storey frame-E1-Interstorey Drift................................................................................................208 Fig. 186. 9-Storey frame-E1-Interstorey Shear ...............................................................................................209 Fig. 187. 9-Storey frame-E1-Interstorey Moment..........................................................................................209 Fig. 188. 9-Storey frame-E2-Floor Displacement ..........................................................................................210 Fig. 189. 9-Storey frame-E2-Interstorey Drift................................................................................................211 Fig. 190. 9-Storey frame-E1-Interstorey Shear ...............................................................................................211

xiii

Fig. 191. 9-Storey frame-E1-Interstorey Moment......................................................................................... 212 Fig. 192. 9-Storey-weak frame-E1-Floor Displacement............................................................................... 213 Fig. 193. 9-Storey-weak frame-E1-Interstorey Drift..................................................................................... 213 Fig. 194. 9-Storey-weak frame-E1-Interstorey Shear.................................................................................... 214 Fig. 195. 9-Storey-weak frame-E1-Interstorey Moment .............................................................................. 215 Fig. 196. 9-Storey-weak frame-E2-Floor Displacement............................................................................... 215 Fig. 197. 9-Storey-weak frame-E2-Interstorey Drift..................................................................................... 216 Fig. 198. 9-Storey-weak frame-E1-Interstorey Shear.................................................................................... 217 Fig. 199. 9-Storey-weak frame-E1-Interstorey Moment .............................................................................. 217 Fig. 200. 9-Storey frame .................................................................................................................................... 220 Fig. 201. 20-Storey frame .................................................................................................................................. 221 Fig. 202. 9-Storey frame .................................................................................................................................... 222 Fig. 203. 20-Storey frame .................................................................................................................................. 223 Fig. 204. 9-Storey frame .................................................................................................................................... 224 Fig. 205. 20-Storey frame .................................................................................................................................. 225 Fig. 206. 9-Storey frame .................................................................................................................................... 226 Fig. 207. 20-Storey frame .................................................................................................................................. 227 Fig. 208. 9-Storey frame .................................................................................................................................... 228 Fig. 209. 20-Storey frame .................................................................................................................................. 229 Fig. 210. 9-Storey frame .................................................................................................................................... 230 Fig. 211. 20-Storey frame .................................................................................................................................. 231 Fig. 212. 9-Storey frame .................................................................................................................................... 232 Fig. 213. 20-Storey frame .................................................................................................................................. 233 Fig. 214. 9-Storey frame .................................................................................................................................... 234 Fig. 215. 20-Storey frame .................................................................................................................................. 235

xiv

xv

List of Tables
Table 1. Nonlinear Static Procedures in earthquake engineering.....................................................................8 Table 2. FEMA 356: lateral load distribution requirements .......................................................................... 30 Table 3. Member properties (from Nakashima, 2006) ................................................................................... 53 Table 4. Material properties after testing .......................................................................................................... 53 Table 5. 9-Ground Motions characteristics...................................................................................................... 57 Table 6. Near-Field records: total drift recorded in dynamic analyses ......................................................... 58 Table 7. SAC joint venture records: PGA and duration................................................................................. 58 Table 8. Correction factor................................................................................................................................. 103 Table 9. Steel section properties....................................................................................................................... 119 Table 10. Building models: first three periods of vibration ......................................................................... 121 Table 11. 3-Storey frame: mode shapes. ......................................................................................................... 121 Table 12. 9-Storey frame: mode shapes. ......................................................................................................... 121 Table 13. 20-Storey frame: mode shapes. ....................................................................................................... 122 Table 14. 3-Storey frame: scaling factors for OGM...................................................................................... 123 Table 15. 3-Storey weak-frame: scaling factors for OGM ........................................................................... 123 Table 16. 9-Storey frame: scaling factors for OGM...................................................................................... 123 Table 17. 9-Storey weak-frame: scaling factors for OGM ........................................................................... 124 Table 18. Total drift recorded in the analyse [%] .......................................................................................... 156 Table 19. 3-Storey frame-drift level 0.5%-E1 (mean) ................................................................................... 174 Table 20. 3-Storey frame-drift level 0.5%-E1 (standard deviation) ............................................................ 174 Table 21. 3-Storey frame-drift level 2%-E1 (mean) ...................................................................................... 175 Table 22. 3-Storey frame-drift level 2%-E1 (standard deviation) ............................................................... 175 Table 23. 3-Storey frame-drift level 4%-E1 (mean) ...................................................................................... 175 Table 24. 3-Storey frame-drift level 4%-E1 (standard deviation) ............................................................... 175 Table 25. 3-Storey weak frame-drift level 0.5%-E1 (mean) ......................................................................... 176 Table 26. 3-Storey weak frame-drift level 0.5%-E1 (standard deviation) .................................................. 177 Table 27. 3-Storey weak frame-drift level 2%-E1 (mean) ............................................................................ 177 Table 28. 3-Storey weak frame-drift level 2%-E1 (standard deviation) ..................................................... 177 Table 29. 3-Storey weak frame-drift level 4%-E1 (mean) ............................................................................ 177 Table 30. 3-Storey weak frame-drift level 4%-E1 (standard deviation) ..................................................... 177 Table 31. 9-Storey frame-drift level 0.5%-E1 (mean) ................................................................................... 179 Table 32. 9-Storey frame-drift level 0.5%-E1 (standard deviation) ............................................................ 179 Table 33. 9-Storey frame-drift level 2%-E1 (mean) ...................................................................................... 179 Table 34. 9-Storey frame-drift level 2%-E1 (standard deviation) ............................................................... 179 Table 35. 9-Storey frame-drift level 4%-E1 (mean) ...................................................................................... 180 Table 36. 9-Storey frame-drift level 4%-E1 (standard deviation) ............................................................... 180 Table 37. 9-Storey weak frame-drift level 0.5%-E1 (mean) ......................................................................... 181 Table 38. 9-Storey weak frame-drift level 0.5%-E1 (standard deviation) .................................................. 181 Table 39. 9-Storey weak frame-drift level 2%-E1 (mean) ............................................................................ 182 Table 40. 9-Storey weak frame-drift level 2%-E1 (standard deviation) ..................................................... 182 Table 41. 9-Storey weak frame-drift level 2.7%-E1 (mean) ......................................................................... 182 Table 42. 9-Storey weak frame-drift level 2.7%-E1 (standard deviation) .................................................. 182

xvi

1
1.1 Forward

INTRODUCTION

In the last two decades, with the development of Performance-Based Seismic Engineering (PBSE) concepts, the demand for the definition of simplified methods for estimating, with an adequate level of confidence, the seismic demand for structures has increased. For the seismic evaluation of yielded systems, inelastic displacements rather than elastic forces should be a more rational approach, since the first directly relate with damage. Previous results (Priestley, 1993) confirmed this statement assessing that traditional Force-Based Design procedures (FBD) lead to unrealistic results. The application of Performance-Based Design (PBD) principles thus requires the definition of analysis procedures able to provide an accurate prediction of such inelastic mechanisms avoiding an excessive computational effort. Among these, Nonlinear Static Procedures (NSPs) appear as one of the most attractive analysis tool since they are simple to use and since they provide a simple and effective graphical representation of the structural response, by means of the so called Pushover Curve. This kind of procedures start from a pushover analysis, identify an equivalent single degree of freedom system, and then estimate the seismic demand for the design response spectrum, usually in terms of roof displacement. Finally, engineering response parameters of interest can be found through the predicted response by the pushover analysis at the design displacement. Whilst NSPs (as coefficient methods or equivalent linearization procedures) differ only in the approach used to estimate the global displacement demand, the pushover method adopted will affect not only the global response demand but also the local response parameters of interest, because both are related to the capacity curve obtained. For this reason, a more accurate prediction of the dynamic response by

Evaluation of Pushover Procedures for the Seismic Design of Buildings

means of a pushover analysis is a fundamental element, and thus the call for further improvements in this field has been increasing in the last few years. In a pushover analysis, a mathematical model of the building, that includes all significant lateral force resisting members, is subjected to a monotonically increasing invariant (or adaptive) lateral force (or displacement) pattern until a pre-determined target displacement is reached or the building is on the verge of incipient collapse. Due to the static nature of the analysis the overall response of the system cannot be reliably estimated principally due to (i) higher mode effects and/or (ii) high ductility demand, which are the main issues investigated in the present work. In particular, the higher mode contributions are typically difficult to identify, and the spreading of inelastic deformations among the structural members leads to degradation and softening of the system resulting in period elongation and change of modal shape characteristics, not accounted for in traditional pushover schemes. Moreover, pushover procedures, due to their static nature, are unable to reproduce peculiar dynamic effects, such as sources of energy dissipation (kinetic energy and viscous damping) as well as duration effects, and account for a sitespecific response by considering both the actual dynamic properties of the system and the frequency content of the seismic motion. Three-dimensional effects are also difficult to incorporate, whereas the effects of cyclic earthquake loading cannot be modelled. Conventional pushovers consist in the application and monotonic increase of a predefined lateral force pattern, kept constant throughout the analysis. The lateral load pattern should approximate the inertial forces expected in the building during an earthquake. However, the inertia force distribution will vary with the severity of the earthquake and with time, due to changes in the contribution of different modes and also as a consequence of the spread of inelastic deformations into the system. Thus the adoption of an invariant load pattern is an approximation that is likely to yield accurate predictions only for low to medium-rise framed structures, where the system behaviour is dominated by a single mode. With the assumption of an adaptive, instead of invariant, lateral load pattern (either forces or displacements), updated knowing the current dynamic properties of the system during the analysis, alteration of the local resistance and modal characteristics of the structure can be accounted for. In this way the stiffness degradation and period elongation induced by the progressive accumulation of damage can be taken into account during the pushover analysis. Moreover, the contribution of several modes can be accounted for, through an appropriate combination rule of their respective shapes. In this way, the

Evaluation of Pushover Procedures for the Seismic Design of Buildings

effect of higher modes can be considered, and if the modal contributions are weighted according to a selected response spectrum the attainment of site-specific results is also possible. Among the adaptive procedures the Displacement-based Adaptive Pushover (DAP) (Antoniou and Pinho 2004(b)) represents an appealing alternative to traditional invariant load shape procedures. The lateral load distribution is continuously updated according to the modal shapes and participation factors derived by eigenvalue analysis carried out at each analysis step. DAP manages to provide greatly improved predictions throughout the entire deformation range, even if an exact reproduction of the dynamic response cannot be achieved (Antoniou and Pinho, 2004(b)). The present work fits within the above described research field, with the main objective to further assess the accuracy of different pushover analysis procedures in predicting the dynamic response of multi-storey structures at different ductility (i.e. drift) demand levels. Traditional (invariant) load shape schemes as well as the recently proposed DAP are considered and their performance compared with respect to nonlinear dynamic analyses.

1.2

Aims of the Work

Nonlinear static pushover analysis, even if fundamentally limited due to its static nature, represents the most attractive alternative to nonlinear response history analysis tools. The main limitations of these analysis procedures are essentially related to the development of higher mode-dominated responses, and the increase of the inelastic deformation demand throughout the structure. Therefore, the main objectives of the present work will be to: Evaluate and compare performances of traditional pushover schemes with those obtained with the more recently proposed Displacement-based Adaptive Pushover (Antoniou and Pinho, 2004(b)). The accuracy of pushover methods will be assessed in predicting (i) the global response, through a comparison of pushover curves with Incremental Dynamic Analysis (IDA) envelopes (Hamburger et al., 2000; Vamvatsikos and Cornell, 2002; Mwafy and Elnashai, 2000), as well as (ii) local response quantities, such as storey displacements, interstorey drifts, interstorey shears and moments. Verify the applicability of an alternative adaptive scheme which makes use, instead of a recordspecific spectrum, of an average design spectrum, as commonly adopted in the current design practice. Further assess (with respect to Antoniou and Pinho (2004) and Lpez-Menjivar (2004)) the effectiveness of the three plot options (maximum total drift vs. correspondent base shear,

Evaluation of Pushover Procedures for the Seismic Design of Buildings

correspondent total drift vs. maximum base shear, maximum total drift vs. maximum base shear) of IDA envelopes. Assess the reliability of DAP for assessment of steel-frame structures, something that has so far not been carried out.

1.3

Thesis Outline

The second chapter presents the current state-of-art of Nonlinear Static Procedures in earthquake engineering: the most updated code provisions are briefly presented so as to highlight the role played by pushover analysis in the current design practice. Then, current pushover procedures are carefully discussed, starting from conventional techniques to the most recent multimodal and adaptive schemes. Moreover, the main drawbacks in current pushover procedures, together with questions regarding the dynamic response of Multiple Degrees Of Freedom Systems (MDOFSs), are pointed out and discussed, in order to highlight the main objectives of the present work. The third chapter is constituted by an overall summary of the performed study. Most important features and assumptions in prototype buildings modeling are described, together with the selection of the employed input ground motions and the fibre-modeling finite element program in which all the analyses have been performed is introduced. Finally, analyses performed and procedures for postprocessing of results obtained are presented. In chapter four results obtained under the point of view of the global response prediction are presented. Pushover curves provided by each pushover method are compared with IDA envelopes, and their performance evaluated. Moreover, the effectiveness of the three plot options (maximum total drift vs. correspondent base shear, correspondent total drift vs. maximum base shear, maximum total drift vs. maximum base shear) of IDA envelopes is further assessed. The fifth chapter shows results obtained under the perspective of local response prediction. Predictions of storey displacements, interstorey drifts, interstorey shears and moments given by static procedures, are compared with nonlinear dynamic analyses responses, and the effectiveness of every pushover method evaluated. Finally, in chapter six concluding remarks are presented and unsolved issues, which might constitute objectives for future research, highlighted.

2
2.1 Introduction

Nonlinear Static Procedures: Critical Overview

In the last few years with the development of performance-based design procedures, the demand for the definition of simplified methods to estimate, with an adequate level of confidence, the seismic demand for structures has increased. For the seismic evaluation of yielded systems, the consideration of inelastic displacements rather than elastic forces should be a more rational approach, because as the structure starts responding in inelastic manner the displacements keep increasing at relatively constant levels of lateral forces. Previous results (Priestley, 1993) have shown that traditional Force-Based Design procedures (FBD) are clearly flawed. Some of the major drawbacks are that (i) they do not account for force redistribution following yielding, and (ii) they do not consider potential failure modes that arise from mid and upper storey mechanisms caused by the influence of higher modes. The application of Performance-Based Design principles (PBD) thus requires the definition of analysis procedures able to provide an adequate prediction of such inelastic mechanisms avoiding an excessive computational effort. It is unquestionable that nonlinear dynamic analysis is the most accurate method for assessing the response of structures subjected to earthquake action. Indeed, any type of static analysis will always be inherently flawed, given the conspicuous absence of time-dependent effects. However, as noted by Goel and Chopra (2005(a)), amongst others, such type of analysis is not without its difficulties or drawbacks, particularly for what concerns application within a design office environment. Firstly, in order to employ dynamic analysis for seismic design/assessment of structures, an ensemble of site-specific ground motions compatible with the seismic hazard spectrum for the site must be

Evaluation of Pushover Procedures for the Seismic Design of Buildings

simulated. As described by Bommer and Acevedo (2004), amongst others, this is, however, a far from simple task, since seismic design codes feature insufficient or inadequate guidance on procedures to either (i) generate artificial spectrum-compatible records, (ii) produce synthetic accelerograms from seismological models or (iii) select appropriate suites of real acceleration time-series, eventually modified to better fit a given code response spectrum. It is believed that until better guidance on record selection/generation will be made available to earthquake engineer designers, this first step will remain as a very difficult-to-overcome hurdle to the use of dynamic time-history analysis in design office applications. Secondly, notwithstanding the significant increase in computing power witnessed in recent years, nonlinear time-history analysis remains computationally demanding, especially when fibre-based (distributed inelasticity) structural analysis programs, which are simpler to calibrate than their plastichinge (concentrated plasticity) counterparts, are employed to model the seismic response of large multistorey irregular buildings, requiring 3D models with thousands of elements. This problem becomes even the more significant if one considers that the analyses will need to be repeated a significant amount of times, not only because design codes or guidance documents request for a relatively large number of earthquake records to be employed in order to warrant minimum probabilistic validity of the results, but also, and perhaps mainly, because the process of analysing any given structure is invariably an iterative one, given that modelling errors are commonly encountered as the design/assessment process evolves. Thirdly, even in those situations where the expertise and resources for running time-history analyses are available, it is often the case that preliminary simpler analysis (i.e. modal and static analyses) are run to enable a first check of the model; errors in the definition/assemblage of a finite elements model are difficult to detect from dynamic analysis results, whilst they tend to be relatively evident from the output of eigenvalue or pushover runs. As an example, inspection of the first modes of vibration of a given building model may be used to check if member and mass has been correctly distributed, whilst examination of a force-displacement monotonic capacity curve may serve to quickly assess if member strength and ductility has been properly assigned. In fact, static analyses, even if representing simplified methods, provide also many important structural response information, such as (i) identification of critical regions, where large inelastic deformations may occur, (ii) individuation of strength irregularities in plan and elevation that might cause important changes in the inelastic dynamic response characteristics (e.g. Krawinkler and Seneviratna, 1998), (iii) evaluation of the force demand in potentially brittle elements, and (iv) prediction of the sequence of yielding and/or failure of structural members. In addition, the explicit insight that pushover-derived base shear vs. top displacement capacity curves provide into the stiffness, strength and ductility of a given structure, constitutes the type

Evaluation of Pushover Procedures for the Seismic Design of Buildings

of qualitative data that is always most informative and useful, within a design application, even when time-history analysis is then employed for the definitive verifications. The above constitute, in the opinion of the author, strong reasons for nonlinear static analysis methods to continue to be developed and improved (as recently proposed documents, such as the FEMA 440, also confirm), so that these tools can become even more reliable and useful when employed either as a replacement to time-history analysis in the seismic design/assessment of relatively simple non-critical structures, or as a complement to dynamic analysis of more complex/critical facilities. It is, therefore, within this framework of warranted development of nonlinear static analysis procedures that the current endeavour finds its justification and rationale.

2.2

NSPs in Earthquake Engineering

Nonlinear Static Procedures appear as one of the most attractive analysis tool due to their ease of use and also because they provide a simple and effective graphical representation of the structural response, by means of the so called Pushover Curve. The latter relates directly the capacity of the system, usually in terms of base shear, with the response of a significant structural node (control node): this kind of representation of the overall response allows for a direct idealization of the system as a Single Degree Of Freedom System (SDOFS) that greatly simplifies the design (or assessment) procedure. The majority of the Nonlinear Static procedures follow the same basic principles: 1. A pushover analysis is performed. 2. An equivalent SDOFS, based on the pushover curve, obtained throughout a static pushover analysis, is defined. 3. The maximum global displacement demand is estimated, according to a selected design response spectrum. 4. The SDOFS response and the actual response of the structure are related by means of a shape coefficient, typically identified in the first mode participation factor. 5. Finally, the response parameters, storey drift and forces on each structural member, can be evaluated, knowing the global demand, through the pushover curve (or capacity curve) of the system. Due to the simplified nature of such methods, they involve many unsolved issues regarding both the capability to capture the dynamic response by means of a pushover analysis as well as the effectiveness of the SDOFS idealization.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

All the proposed NSPs differ essentially in the definition of the global displacement demand (step 3) and can be classified in two main groups: equivalent linearization procedures and coefficient methods (Table 1). Table 1. Nonlinear Static Procedures in earthquake engineering Rosenblueth and Herrera (1964) Glkan and Sozen (1974) Equivalent linearization procedures Iwan (1980) DDBD, Priestley and Kowalsky (2000) CSM, Freeman (1994) Improved CSM (Chopra and Goel, 1999) Newmark and Hall (1982) Coefficient methods Miranda (2000) FEMA 356 (ASCE, 2000) N2-method, Fajfar (1999) Hereafter, since the work is not focused on the evaluation of a single NSP methodology, only the most updated code procedures, which make use of a pushover-based analysis, are briefly summarized for clarity: i.e. the Eurocode 8 (prENV 1998-1, 1994), FEMA 356 (ASCE, 2000), ATC-40 (ATC, 1996) and FEMA 440 (ATC, 2005).
2.2.1

The Eurocode 8 (prENV 1998-1, 1994)

The NSP adopted in this seismic design code is the N2-procedure developed by Fajfar (1999) which consists in the definition of a bilinear SDOFS (Fig. 1) corresponding to the first vibration mode assuming that the contribution of the other modes is negligible. The design displacement (or maximum displacement response) for the design earthquake is defined as the displacement response spectrum at the elastic period of the SDOFS (T*) accounting for the system ductility, by means of an amplification factor, whereas the equivalent displacement approach between the linear system response and the response of the nonlinear one cannot be applied (i.e. short period range).

Evaluation of Pushover Procedures for the Seismic Design of Buildings

Fig. 1. Equivalent Single Degree Of Freedom System

2.2.2

The Coefficient Method in FEMA 356 (ASCE, 2000)

The NSP adopted in FEMA 356, the coefficient method, consists in the definition of an equivalent linear SDOFS considering an effective period Te generated from the initial period Ti, accounting for some loss of stiffness in the transition from the elastic to inelastic behaviour. This procedure estimates the total maximum displacement of the SDOF oscillator by multiplying the elastic SDOFS response (assuming the initial linear properties, stiffness and damping) by one or more coefficients empirically derived (Fig. 2). These coefficients accounts for (i) the SDOF idealization (a shape factor scales the SDOFS response to the roof displacement of the building), (ii) the linear response assumed (conventionally characterized in terms of strength, ductility and period (R--T relationships)), (iii) stiffness and strength degradation, and (iv) the dynamic amplification of the response due to P- effects. It might be noted that the design displacement is defined by means of an iterative procedure until convergence of the linear SDOFS displacement amplitude to the response spectrum ordinate.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

10

Fig. 2. Coefficient Method (ASCE, 2000)

2.2.3

The Capacity Spectrum Method in ATC 40 (ATC, 1996)

The NSP adopted in this code is the capacity spectrum method proposed by Freeman (1994). This technique, following an equivalent linearization approach, estimates the maximum global displacement of the structure by means of an iterative graphical procedure (Fig. 3). The basic assumption is that the maximum inelastic deformation of a nonlinear SDOFS can be approximated from the maximum deformation of a linear elastic SDOFS with a larger period and damping ratio than the initial values of the inelastic one. According to this procedure the capacity curve is converted into an equivalent SDOFS pushover response and plotted on the same axes as the seismic ground motion demand in the AccelerationDisplacement Response Spectrum (ADRS) format, assuming a trial damping ratio. The secant period at the interception identifies the equivalent period of the elastic SDOFS with an equivalent viscous damping ratio proportional to the area enclosed by the capacity curve of the equivalent nonlinear SDOFS. Since both the period and damping are function of the displacement, the procedure requires iterations until the assumed damping is equal to the value computed at the design displacement.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

11

Fig. 3. Capacity Spectrum Method (ATC, 1996) Recently, the ATC-55 projects (ATC, 2005) demonstrate that the Coefficient Method (as proposed in FEMA 356) as well as the CSM (as proposed in ATC-40) show some inconsistencies in the prediction of the displacement demand. Thus they propose a new formulation of both design approaches. The updated design methods obtained lead to approximately the same results with a significant improvement in the prediction of the displacement demand when compared to response history analysis results. As it might be concluded from the discussion reported above, these procedures differ only in the approach used to estimate the global displacement demand (global response parameter, i.e. top floor displacement or the equivalent SDOFS displacement demand); instead, the pushover method adopted will affect not only the global response demand but also the local response parameters of interest, because both are related to the capacity curve obtained. For this reason a more accurate prediction of the dynamic response by means of a pushover analysis is a fundamental element, and thus the call for further improvements in this field was increase in the last few years.

2.3

Pushover Analysis in Earthquake Engineering

Pushover analysis, which represents the fundamental analytical tool in all the NSPs quoted above, is a static method that directly incorporates nonlinear material characteristics. A mathematical model of the building, that includes all significant lateral force resisting members, is subjected to a monotonically

Evaluation of Pushover Procedures for the Seismic Design of Buildings

12

increasing invariant (or adaptive) lateral force (or displacement) pattern until a pre-determined target displacement is reached or the building is on the verge of incipient collapse. The term pushover analysis describes a modern variation of the classical collapse analysis method, as fittingly described by Kunnath (2004). It refers to an analysis procedure whereby an incremental-iterative solution of the static equilibrium equations is carried out to obtain the response of a structure subjected to monotonically increasing lateral load pattern. The structural resistance is evaluated and the stiffness matrix is updated at each increment of the forcing function, up to convergence. The solution proceeds until (i) a predefined performance limit state is reached, (ii) structural collapse is incipient or (iii) the program fails to converge. In this manner, each point in the resulting displacement vs. base shear capacity curve represents an effective and equilibrated stress state of the structure, i.e. a state of deformation that bears a direct correspondence to the applied external force vector. Even if representing a simplified analytical tool, with respect to nonlinear dynamic analyses, they may provide important structural response information such as: Identify the progress of the overall capacity curve of the structure. Identify critical regions, where large inelastic deformations may occur. Identify strength irregularities in plan and elevation that might cause important changes in the inelastic dynamic response characteristics (e.g. Krawinkler and Seneviratna, 1998). Estimate the force demand in potentially brittle elements. Predict the sequence of yielding and/or failure of structural members.

Many authors (Lawson et al., 1994; Krawinkler and Seneviratna, 1998; Kim and DAmore, 1999; Naeim and Lobo, 1999; Antoniou and Pinho, 2004(a)) have however noted that nonlinear static methods suffer from a number of limitations, which stem essentially from their inherently static nature. Such limitations become particularly evident when high-rise flexible frames, whose response may be heavily influenced by higher modes, are being assessed. The most important limitations in the applicability of current NSPs are due to the static nature of this kind of analysis. In fact NSPs assume that all the structural response quantities (displacements, internal forces, plastic deformations etc.) can be estimated by means of those recorded in a pushover analysis at the design displacement level. Due to the static nature of the analysis, the overall response of the system cannot be reliably estimated because of an inaccurate deformation prediction, principally due to (i) higher mode effects and/or (ii) high ductility demand. Both these contributions, which are the main issues investigated in the present work, lead to a redistribution of the internal forces with respect to the commonly assumed 1st mode

Evaluation of Pushover Procedures for the Seismic Design of Buildings

13

response, which results in a concentration of deformations at the location of damage. In particular, the higher mode contributions are typically difficult to identify, and the spreading of inelastic deformations among the structural members leads to degradation and softening of the system resulting in period elongation and change of modal shape characteristics. Moreover, pushover procedures are unable to reproduce peculiar dynamic effects, such as: (1) consider not only the strain energy, but also other sources of energy dissipation (kinetic energy and viscous damping) as well as duration effects; (2) account for a site-specific response by considering both the actual dynamic properties of the system and the frequency content of the seismic motion. Three-dimensional effects are also difficult to incorporate, whereas the effects of cyclic earthquake loading cannot be modelled. With the assumption of an adaptive, instead of an invariant, lateral load pattern (either forces or displacements), updated knowing the current dynamic properties of the system during the analysis, alteration of the local resistance and modal characteristics of the structure can be accounted for. In this way the stiffness degradation and period elongation induced by the progressive accumulation of damage can be taken into account during the pushover analysis. Further improvements, as discussed later, can be achieved considering the contribution of several modes by combining their respective responses with an appropriate combination rule. In this way the effect of higher modes can be considered, and if the modal contributions are weighted according with a selected response spectrum site-specific results can also be obtained. Finally, an important issue, currently ignored in all code implementations, regards the definition of a displacement load vector rather than lateral forces. Although it cannot constitute an acceptable approach using an invariant vector shape since it leads to a predefined and often unreliable failure mode (i.e. deformed shape), it allows for a better understanding of the deformation of the structure in the inelastic regime if an adaptive approach is used (Antoniou and Pinho, 2004(b)).
2.3.1 Non-Adaptive Non-Modal Procedures (NANM Procedures)

Conventional pushover consists in the application and monotonic increase of a predefined lateral force pattern, kept constant throughout the analysis. The lateral load pattern should approximate the inertial forces expected in the building during an earthquake. However, the inertia force distribution will vary with the severity of the earthquake and with time, due to changes in the contribution of different modes in the elastic range and also as a consequence of the spread of inelastic deformations into the system. Thus, the adoption of an invariant

Evaluation of Pushover Procedures for the Seismic Design of Buildings

14

load pattern is an approximation that is likely to yield accurate predictions only for low to medium-rise framed structures, where the system behaviour is dominated by a single mode. Usually the following NANM lateral load force vectors are adopted: 1st Mode: forces are proportional to the amplitude of the elastic first mode shape and mass at each floor. Inverted Triangular: forces are linearly increasing with height, in order to approximate a 1st mode shape vector for a frame structure with regular geometric and mass distribution. Uniform (or Rectangular): the force pattern is uniform with the building height. It is an approximate representation of the inertial forces distribution in the inelastic range assuming the structure has formed a soft storey. Code distribution: code-specified distribution that usually varies from an inverted triangular to a parabolic shape. Many researchers have analysed the performances and defined the limitations of using this kind of force-patterns. Among these Mwafy and Elnashai (2000), and Gupta and Kunnath (2000), have found that whereas in the elastic range force distributions with a triangular or a trapezoidal shape provides a better fit to dynamic analysis results, at large deformations, after that the structure has sustained significant damage at a particular storey level, the dynamic envelopes are closer to the uniformly distributed force solutions.
2.3.2 Non-Adaptive Modal Procedures (NAM Procedures)

This kind of procedures try to improve the performances of the static methods by focusing the attention on the higher modes effects, by considering several modes contribution in the definition of the applied lateral load pattern. Among these, an SRSS (Square Root of the Sum of the Squares) load vector is considered in the ATC55 project (ATC, 2005). The lateral force vector is defined as the vector that produces an interstorey shear distribution correspondent to the SRSS combination of the storey shears obtained with different modal force distributions. As shown in FEMA 440 (ATC, 2005), however, it leads to inconsistent improvements with respect to other NANM load vectors (e.g. 1st Mode vector) requiring also greater computational effort.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

15

Recent years have also witnessed the development and introduction of an alternative type of nonlinear static analysis, which involves running multiple pushover analyses separately, each of which corresponding to a given modal distribution, and then estimating the structural response by combining the action effects derived from each of the modal responses (i.e. each displacement-force pair derived from such procedures does not actually correspond to an equilibrated structural stress state). Paret et al. (1996) and Sasaki et al. (1998) suggested the Multi-Mode Pushover analysis (MMP). A MMP analysis consists in performing several pushover analyses under forcing vectors representing the various modes deemed to be excited in the dynamic response. For each pushover curve obtained, the Capacity Spectrum Method (ATC, 1996) is utilized to compare the structural capacity with the earthquake demand. This method leads only to the identification of the critical mode but does not provide any results on the overall response of the system since it does not consider any kind of modal combination rule. A refinement of the MMP procedure is given by the Pushover Results Combination (PRC), which was proposed by Moghadam and Tso (2002). According to this method, firstly the structural response parameters related to each mode contribution have to be found according with the MMP approach, and then the overall response is calculated by combining them according with the modal participating factor of each mode. Usually the first three or four modes are considered. Recently, a Modal Pushover Analysis (MPA) procedure has been proposed by Chopra and Goel (2002). Pushover analyses are conducted independently in each mode, using lateral force profiles that represent the response in each of the modes considered. The pushover curve associated with each modal pushover analysis is idealised as the response curve of a bilinear SDOF system (Fig. 4) and response values are determined at the target displacement (computed by a dynamic analysis of the nonlinear SDOFS, or by applying displacement modification or equivalent linearization procedures). Finally responses are combined together using the SRSS method in order to define the overall response. The present methodology has been further improved including P- effects due to gravity loads in the pushover analysis for all modes and computing the plastic hinge rotation directly from the total storey drift values (Goel and Chopra, 2004).

Evaluation of Pushover Procedures for the Seismic Design of Buildings

16

Fig. 4. Properties of the nth mode inelastic SDOF system derived from the corresponding pushover curve (after Chopra and Goel, 2001) The main approximations of this method are that (i) the coupling among modal coordinates due to yielding of the system is neglected (i.e. it is assumed that when the excitation is proportional to the jth mode, the response in that mode is still predominant), and that (ii) the SRSS combination rule is adopted to find the structural response and thus equilibrium cannot be provided. Whilst the first source of approximation relates with the level of the inelastic demand on the system, the second depends mainly on how much the higher modes contribute to the overall response. Moreover, further limitations consist in the fact that (iii) the modal responses that are being combined correspond to different stress states of the same structure since different levels of inelasticity are attained in each mode, and that (iv) the equivalent SDOF system properties depends on the estimated displacement demand and thus an iterative strategy is required to avoid further sources of uncertainties. Chopra and Goel (2002) found that this approach leads to good estimates of displacements and drifts but did not satisfactorily estimate plastic hinge rotations for some high-rise structures (e.g. the 9-storey steel frame considered in this work). Moreover, this method is not able to provide a reasonable estimate of drifts for the upper storeys of tall frames or at large degrees of inelasticity. Further researches (Chintanapakdee and Chopra, 2003), considering several buildings with different strength levels (correspondent to different strength reduction factors for the equivalent SDOFS, i.e. different ductility levels), shown that MPA fails to predict the drifts profile, in particular at the upper storey locations, as the building height as well as the ductility level increases. Goel and Chopra (2004) further developed and tested this procedure for six SAC buildings subjected to ensembles of 20 ground motions. Results indicate that, if the response of the structure is nearly elastic then the main source of error is the SRSS combination rule adopted, whilst, with higher level of

Evaluation of Pushover Procedures for the Seismic Design of Buildings

17

ductility demand, also a higher contribution of P- effects due to gravity loads and a large coupling of the modal responses further affects the degree of accuracy of the pushover prediction. For these reasons, even if the MPA estimates of demand were much better than those provided by the force distribution specified in FEMA-356, they cannot be considered satisfactory for buildings that are deformed well into the inelastic range with significant degradation in lateral capacity. For such cases a nonlinear response history analysis is recommended by Chopra and Goel (2004). In the recently proposed FEMA 440 document (ATC, 2005) this method has been compared with other types of pushover schemes. Results shown that, although often improved over the simple mode load vectors, estimates of interstorey drift over the full height of the buildings may not be consistently reliable. The MPA procedure appear fundamentally limited because its accuracy depends upon the parameter of interest (forces, deformations, plastic hinge rotations, etc.), the characteristics of the structure, and the details of the specific procedure (as other researches confirm (e.g. Yu et al. (2001), Lpez-Menjivar (2004)) and thus it is suggested only as an alternative method to obtain results to compare. For the prototype frame models considered in the ATC-55 project (ATC, 2005) a modified version of the MPA has been adopted, where the first three modes contribution are considered, but the 2nd and 3rd mode equivalent SDOFSs are assumed linear because of the reversal of some higher mode capacity curves. This fact, as pointed out also in (Hernndez-Montes et al., 2004; Goel and Chopra, 2005(b)) represents one of the main drawbacks in the MPA procedure, which thus remains limited and it does not allow an easy application within a design office environment. The present methodology has been further improved for the calculation of member forces (Goel and Chopra, 2006) and also extended to the case of 3D pushover analysis (Chopra and Goel, 2006). However, as discussed by the authors themselves, the method does not always provide satisfactory results for high-rise steel frames.

The above multi-modal procedures constitute a significant improvement with respect to conventional pushover analysis since they explicitly consider the response of more than one vibration mode and the influence of the expected ground motion. However, none of the invariant force distributions accounts for the change in the modal contributions as well as the redistribution of inertia forces due to structural yielding. To overcome these limitations, adaptive procedures, whereby the load vector shape is updated at every single analysis step to reflect the actual dynamic properties of the system (i.e. stiffness state), will represent an attractive alternative.
2.3.3 Adaptive Procedures

Evaluation of Pushover Procedures for the Seismic Design of Buildings

18

Non-Adaptive Non-Modal as well as NAM procedures, even if they represent an improvement with respect to the former, are not theoretically robust because they neglect any change in the vibration properties of the structure due to the penetration in the inelastic range. The main reason behind the underperformance of these conventional pushover methods is the fact that they do not account for the effect that damage accumulation, induced by the increasing deformation levels imposed on the structure, has on the response of the latter. Cumulative material straining introduces a reduction in stiffness which, in turn, causes an elongation of the periods of vibration, which then, depending on the shape of the response spectrum being considered (or on the frequency content of an input record), may trigger significant changes in the response characteristics of the buildings (as clearly represented in Fig. 5 and Fig. 6). Krawinkler and Seneviratna (1998) summarised the above with a single statement: fixed load patterns in pushover analysis are limiting, be they first modal or multimodal derived, because no fixed distribution is able of representing the dynamic response throughout the full deformation range.

6 first mode second mode third mode

Fig. 5. Periods of vibration of a 4-Storey structure under increasing levels of deformation

period (sec)

0 0.0%

0.5%

1.0%

1.5%

2.0%

total drift

Evaluation of Pushover Procedures for the Seismic Design of Buildings

19

12 10 8 6 4 2 0 0.0% top drift = 0.5% top drift = 1.0% top drift = 1.5% top drift = 2.5%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

4.0%

Fig. 6. Interstorey drift profile of a 12-Storey building subjected to increasing levels of deformation Because of the aforementioned limitations, recent years have witnessed the development and introduction of so-called Adaptive Pushover methods whereby the loading vector is updated at each analysis step, reflecting the progressive stiffness degradation of the structure induced by the penetration in the inelastic range, as schematically shown in Fig. 7, below; it might be noted that in adaptive pushover the response of the structure is computed in incremental fashion, through a piecewise linearization procedure (as described in 2.3.5), hence rendering it possible to use the tangent stiffness at the start of each increment, together with the mass of the system, to compute modal response characteristics of each incremental pseudo-system through elastic eigenvalue analysis, and use such modal quantities to congruently update (i.e. increment) the pushover loading vector. .

base shear (kN)

5000 4000 3000 2000 1000 0 0%

1%

2%

total drift

3%

Fig. 7. Adaptive pushover: shape of loading vector is updated at each analysis step Several researchers have proposed adaptive load (forces or displacements) distributions that attempt to follow more closely the time-variant distributions of inertia forces, associated with new modal

Evaluation of Pushover Procedures for the Seismic Design of Buildings

20

properties of the system (Reinhorn, 1997; Bracci et al., 1997; Requena and Ayala, 2000; Albanesi et al, 2002; Gupta and Kunnath, 2000; Elnashai, 2001; Antoniou and Pinho, 2004(a)(b); Aydinoglu, 2003). Such methods account for (i) the site-specific spectrum scaling and (ii) the higher modes contributions, as the NAM schemes did, and also for (iii) the alteration of local resistance and modal characteristics of the structure induced by the accumulation of damage. The solution most commonly adopted is to push the structure with an adaptive lateral force pattern resulting from the combination of force patterns associated with different modes. Applying forces rather than displacements represent a large source of approximation, even in the elastic range, because, due to the sign loss (associated with SRSS or CQC combination rules), the modal contributions can only add to the global force pattern, and thus induce errors in the estimation of local deformations. This problem has been recognised by Aydinoglu (2003) and Antoniou and Pinho (2004(b)), who proposed alternative displacement-based approaches, as it will be discussed later. Moreover, in tandem with the present drive for performance-based seismic engineering, there is also currently a thrust for the development and code implementation of displacement or, more generally, deformation-based design and assessment methods. Therefore, it would seem that applying displacement loading, rather than force actions, in pushover procedures would be an appropriate option for nonlinear static analysis of structures subjected to earthquake action. However, due to the unvarying nature of the applied displacement loading vector, conventional (non-adaptive) displacement-based pushover analysis can conceal important structural characteristics, such as strength irregularities and soft storeys, should the displacement pattern adopted at the start of the analysis not correspond to the structures post-yield failure mechanism. Consequently, when only non-adaptive static nonlinear analysis tools are available, as has been the case throughout the past, force-based pushover does constitute a preferable choice over its displacement-based counterpart. On the other hand, however, if one is able to apply displacements, rather than forces, in an adaptive fashion, that is, with the possibility of updating the displacement loading pattern according to the structural properties of the model at each step of the analysis, then a conceptually appealing deformation-based nonlinear static analysis tool is obtained. Finally, among the adaptive methodologies (be they force- or displacement-based) should be distinguished between single-run procedures, where a single pushover analysis is carried out through the application of an incrementally increasing lateral force pattern, and multiple-run procedures, where several pushover analyses are performed under different forcing vectors and then results combined. An overview of such kinds of pushover schemes will be presented in the following sections.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

21

2.3.3.1 Single-run adaptive pushover analyses

Reinhorn (1997) and Bracci et al. (1997) were the first to introduce a procedure that utilises fully adaptive patterns. The analysis starts by assuming an initial lateral load distribution, usually triangular, whereas the additional loads imposed in subsequent increments are calculated from the instantaneous storey shear resistances of the previous load step.
j Fi j Fi j 1 j +1 Fi Fi j +1 = V j + P V j V j 1 V j

where i and j represent the storey location and analysis step respectively, Fi j +1 is the incremental ith storey force at step j+1, V is the total base shear, F the applied storey force, P j +1 is the incremental base shear. This procedure has been implemented in the dynamic analysis package IDARC (Valles et al., 1996) leading to the attainment of apparently promising results. Lefort (2000) implemented an extended version of this method, which employed an additional force scaling equation to account for higher mode contributions, obtaining, however, limitedly-accurate response predictions. Another method was proposed by Requena and Ayala (2000), who discussed two variations of adaptive pushover (referred to as approaches 2-A and 2-B) and compared them with a modal fixed-pattern scheme. Whilst in one of the approaches the storey loads were derived through an SRSS combination of modal forces, in the other one the lateral loads distribution is defined according to an equivalent fundamental mode, which is defined through a combination of several modes shape using the SRSS rule. This procedure is repeated whenever the structural stiffness changes, as new plastic hinges form and develop, hence the lateral load distribution reflects the current state of inelasticity. Although both the suggested procedures are indeed appealing since they are theoretically rigorous and they explicitly account for higher modes and spectral contributions, analytical results presented by Requena and Ayala (2000) were only limited and the accuracy of the procedures could not be effectively assessed or judged. An alternative adaptive pushover methodology has been recently proposed by Albanesi et al. (2002), who suggested an Adaptive Energy-Based Pushover Analysis (AEPOA) whereby the imposed lateral force/displacement profiles at each step supposedly take into account not only the inertial properties of the structure but also the kinetic energy that the latter is expected to mobilise when subjected to earthquake loading. The results obtained, however, did not seem particularly exceptional compared to conventional pushovers. The AEPOA method were very unstable and had the tendency to stop at very low deformation levels (less than 1% in some cases), whereas the predicted values of the base shear

Evaluation of Pushover Procedures for the Seismic Design of Buildings

22

strength of the building were unreliable. In addition, no clear procedure for applying and updating the lateral loads was described. Finally, Elnashai (2001) proposed an adaptive pushover scheme that seemed to encompass, within a single analysis pushover algorithm, all advanced features described above. This single-run procedure is fully adaptive and multi-modal and accounts for system degradation and period elongation by updating the force distribution at every step (or at predefined steps) of the analysis. The dynamic properties of the structure are determined by means of eigenvalues analyses that consider the instantaneous structural stiffness state, at each analysis step. Site- or record-specific spectral shapes can also be explicitly considered in the scaling of forces, so as to account for the dynamic amplification that expected ground motion might have on the different vibration modes of the structure. The work of Elnashai, however, was confined to the application to simplified, not-necessarily realistic, stick-models. This Force-based Adaptive Pushover (FAP) algorithm was further developed and tested by Antoniou and Pinho (2004(a)), who also proposed and verified an alternative Displacement-based Adaptive Pushover (DAP) (Antoniou and Pinho 2004(b)). Since both procedures follow an analogous theoretical approach and the further extension and verification of the DAP procedure to the case of steel frames constitute the objective of the present work, the adaptive pushover algorithm is presented and discussed in detail in 2.3.5.
2.3.3.2 Multiple-run adaptive pushover analyses

An adaptive multiple-run pushover methodology was proposed by Gupta and Kunnath (2000), in which the applied loads are constantly updated, depending on the instantaneous dynamic characteristics of the structure, and a site-specific spectrum can be used to define the loading pattern. According to that method, eigenvalue analysis is carried out before each load increment, and than different load patterns (Fij), correspondent to the number of modes of interest, are considered. Base shears Vj associated with each mode response are combined with SRSS to find a global value V and then each force pattern is scaled to define their single step increment (SnVj), accounting for the number of steps (Ns) and the target shear (VB) assumed.

Vj =

nStoreys

ij

( 2-1) ( 2-2) ( 2-3) ( 2-4)

V = SRSS (V j )
Sn = VB N sV

V j = S nV j

Evaluation of Pushover Procedures for the Seismic Design of Buildings

23

Then, a static analysis is carried out for each mode independently and the obtained responses for each mode are combined with SRSS and added to corresponding values from the previous step. The estimates of interstorey drifts and the sequence of the formation of local or global collapse mechanisms (i.e. plastic hinges location) presented in the paper were satisfactory. However, the employment of SRSS to combine the different pushover responses, derived for each mode, implies that structural equilibrium is not satisfied at the end of each step. Among the multiple-run adaptive procedures, even if it does not properly belong to the category of pushover-based methods, has to be mentioned the Incremental Response Spectrum Analysis procedure (IRSA) proposed by Aydinoglu (2003). The author considers that the structural response cannot be properly evaluated by means of a NSP since it is based on the definition of a pushover curve and thus it is basically restricted with a single-mode response. Moreover, he recognised that (i) the loading characteristics cannot be based on the elastic instantaneous spectral acceleration, not compatible with the inelastic instantaneous response, that (ii) the contribution of several modes in the plastic hinge formation cannot be neglected, and that (iii) the application of the modal combination in defining the equivalent seismic loads (i.e. actions, rather than combining responses) may lead to un-realistic results. In order to overcame such limitations, the IRSA has the main purpose, by means of a piece-wise linear structural response idealisation and following a displacement-based adaptive methodology, to estimate the structural response quantities of interest without reproducing any pushover curve. Thus it can be treated, more likely, as an alternative procedure with respect to the typically used pushover curvebased NSP. The IRSA follows these basic stages:

First the modal properties of the system are updated at the beginning of each analysis step. The modal capacity diagram (modal acceleration vs. modal displacement) obtained in the previous steps is converted into a bilinear diagram, and than (either by solving the SDOF dynamic equation or by means of a specified response spectrum) the correspondent peak inelastic modal displacement at the current displacement increment i (S(i)din) is calculated.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

24

Fig. 8. Bilinear idealization of a modal capacity diagram at the ith pushover step (Aydinoglu, 2003)

The modal response parameters are combined (using SRSS or CQC rule) and the first modal displacement increment that leads to the formation of a plastic hinge is calculated. Since all the other modal displacements are expressed in terms of the modal displacement increment of the first mode, they can be directly obtained from the latter.

Check if the first mode spectral displacement exceeds the first-mode spectral displacement previously calculated (i.e. one of the modal displacements exceeds the modal inelastic spectral displacement).

If exceeded, the peak response has been reached somewhere, then the modal displacements are set equal to the inelastic spectral displacements for all modes, and the structural response for the design ground motion defined from the combination of the modal responses.

If not, the stiffness matrix has to be updated accounting for the plastic hinge formation and the procedure should start again for the next pushover step until the attainment of the structural capacity.

Even if the preliminary of IRSA results appear to be promising in terms of prediction of the nonlinear structural response quantities, its accuracy needs to be evaluated through statistical studies using different structural systems and earthquake ground motions.
2.3.4 Discussion

Evaluation of Pushover Procedures for the Seismic Design of Buildings

25

Hereafter will be underlined the main drawbacks and open issues that involve both non-adaptive as well adaptive methodologies in order to identify major research topics in the field of pushover analysis procedures.
2.3.4.1 Selection of the control node

In addition to the problem regarding the selection of the lateral load pattern to employ in the pushover analysis, have to be mentioned questions regarding the choice of the roof level as control point (which is strictly related with the idealisation of the MDOF structural response by means of a SDOF system), as pointed out by Hernndez-Montes et al. (2004). Rather than viewing pushover analyses from the perspective of roof displacement, an arbitrary assumption that may lead to a distortion of the capacity curve by means (i) of a disproportionate increase in the roof displacement with respect to the other floors, and even (ii) outright reversal in the case of higher modes pushover analyses (as occurred in the application of MPA in the ATC-55 project (ATC, 2005)), an energy-based displacement is proposed. The latter is defined by considering the work done by the lateral forces acting through the floor displacements (i.e. the absorbed energy) during the pushover analysis. The energy-based pushover procedure suggested by the authors may be implemented both with an invariant or adaptive load vector. This procedure has been tested as a part of the ATC-55 project (ATC, 2005): the energy-based capacity curves are able to correct the anomalies observed for higher mode pushover curves, and in some cases the results obtained represent an improvement with respect to the current methods, but the estimates of response quantities still are not consistently reliable.
2.3.4.2 Alternative modal combination rules

By definition every multi-mode pushover procedure depends on the combination rule adopted in order to define the overall response of the structure. The attempt is to capture the absolute response maxima recorded in a nonlinear dynamic analysis by means of an easier and more direct analytical tool as a pushover procedure represents. Several researchers recognised this problem (among these, Fishingher et al., 2004) and proposed different alternative combination rules (Matsumori et al., 1999; Kunnath, 2004; Antoniou and Pinho, 2004; Lpez-Menjivar, 2004) by means of algebraic summation of mode contributions through positive and negative weighting coefficients. Matsumori et al. (1999) proposed two patterns of storey shear distributions, given respectively by the sum and the difference of the storey shears given by the first two modes. The application of the methodology on two building models produced good correlation with time-history responses.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

26

Lpez-Menjivar (2004) developed the application of a Direct Vectorial Addition (DVA) procedure to both the FAP and DAP algorithms. Two suitable arrangements of such scheme has been obtained: positive contribution of the first and the third modes, and negative of the second in FAP, while positive contribution of the first mode, negative 10% contribution of the second mode, and positive 10% contribution of the third mode, for DAP. Such modal combination rules have been defined in order to provide a satisfactory estimate of the dynamic interstorey drift profile for a set of seismically designed structures. However, even if in its displacement-based variant DVA always represents an improvement with respect to quadratic combination rules in the prediction of the dynamic drift trend, results for different structural systems shown that the accuracy provided by the DVA methodology is sensitive to the structural model analysed and it is not always satisfactory. Moreover, the nature of these weighting factors is arbitrary and thus until a general procedure to correctly determine these values is found the DVA adaptive pushover modality cannot really be deemed as a valid solution for practical application. The modal combination procedure proposed by Kunnath (2004) is based on the computation of the response quantities by an envelope of those obtained throughout several non-adaptive pushovers, with lateral load shape correspondent to all the possible combinations among the number of modes considered. Moreover, different weighting coefficients (i) for each mode contribution can be introduced. Where the first three modes contribution, suggested adequate to obtain conservative results, are included the following combination can be used. F j = 11 m 1 S a (1 , T1 ) 2 2 m 2 S a ( 2 , T2 ) 3 3 m 3 S a ( 3 , T3 ) ( 2-5)

This method provides a good representation of interstorey drift and interstorey shear along the height of the building. However, the method remains fundamentally limited due to the fact that the selection of the weighting coefficients and the number of modes to include in the analysis are an arbitrary choice of the designer. Moreover, several pushover runs, using alternative lateral load patterns, are required to identify an envelope of possible outcomes representative of the structural response. Moreover, it might be noted as any combination rule seems to give satisfactory results also because of limitations inherent to the static nature of pushover analysis procedure, that can only account in approximate way of dynamic effects:

Two kind of envelopes have to be accounted for: (i) Design response maxima (i.e. envelope of the dynamic response), which are the main design objectives, occur at different times during a response-history analysis (time envelopes); as highlighted by Gupta and Kunnath (2000), by

Evaluation of Pushover Procedures for the Seismic Design of Buildings

27

analysing the recorded response of instrumented buildings, they found that the inertia force pattern not only is subjected to a great variation at various instants of time in the response history, but also has an extremely different shape at the instant when each response parameter (floor displacement, interstorey drift, shear or moment) reaches its peak value. Moreover, (ii) they are envelopes among different modal contributions (modal envelopes).

In addition the definition of a rational method to combine different mode contributions there is the choice between a solution that satisfies or not equilibrium. Since it is impossible to predict both the dynamic response envelope and the current time-step deformed shape, the assumed modal combination rule will inevitably favour the former, which implies solutions that violate equilibrium, or the latter, where equilibrium have to be satisfied. Whilst algebraic addition methods represent solutions that satisfy equilibrium conditions, probabilistic approaches (e.g. SRSS, CQC) violate this last requirement.

Until the definition of a more rational approach, algebraic addition methods appear too much dependent on the structural model and the characteristics of the earthquake motion, as well as their interaction.

Hence, none of direct algebraic summation schemes can be assesses to be fully theoretically justified and a probabilistic approach (e.g. CQC) seems now the more appropriate choice and thus it was selected in all the analyses of the present research.
2.3.4.3 Drawbacks in multiple-run analyses

As highlighted by their respective authors, the main advantage of multiple-run static analysis procedures is that they may be applied using standard readily-available commercial software packages. The associated drawback, however, is that these methods are inevitably more complex than running a single pushover analysis as noted by Maison (2005). Furthermore, some of the latter multiple-run procedures, either they have an adaptive (Gupta and Kunnath, 2004; Aydinoglu, 2003) or non-adaptive (Paret et al., 1996; Sasaki et al., 1998; Moghadam and Tso, 2002; Chopra and Goel, 2002) nature, will inevitably lead to difficulties when applied within a capacity-spectrum type of procedure due to the need to handle capacity curves associated with higher mode force patterns that display a reversal of the roof displacement as inelasticity develops in the structure (Hernndez-Montes et al. 2004; ATC 2005; Goel and Chopra 2005(b)). For all the above, these multiple-based pushover based approaches do not constitute the scope of the current work, where focus is placed instead on single-run pushover analysis procedures.
2.3.4.4 Multiple Degrees Of Freedom Effects (MDOFEs)

Evaluation of Pushover Procedures for the Seismic Design of Buildings

28

The dynamic response of high-rise frames cannot be easily estimated because traditional pushover schemes are based on a constant-shape load vector. Such force pattern influences the distribution of localised demand in a multiple degrees of freedom (MDOF) system (interstorey drift, interstorey shear, etc.). However, this distribution of forces, assumed constant throughout the pushover analysis, will change during the actual response of the system for a given seismic motion, due to changes in the contribution of the different modes in the elastic range (affected by the dynamic characteristics of the ground motion), and also as a consequence of the spread of inelastic deformations into the system. Typically, higher mode effects on the overall structural response typically involve:

Changes in the inertia force pattern, with respect to the widely adopted in the common practice first mode and triangular shapes, involving possible sign reversals of storey forces. Alteration of the storey drift distribution: whilst it is essentially uniform for low-rise buildings, it became irregular for taller frames. In fact, at the upper storeys of elastic high-rise frames (where higher modes contribution increases) larger drifts are expected. Thus the maximum drift might not occur at the first storey, as the traditional approaches will predict.

Changes of the yield mechanism (i.e. location of plastic hinges) and possible development of upper storey mechanisms.

Moreover, spreading of inelastic deformation among the structural members, correspondent to large ductility and total drift levels, yields:

A redistribution of forces following yielding. Gradual degradation and softening of the system induce a concentration of deformations at locations that sustain more damage and a correspondent change of modal shape characteristics. Cumulative material straining introduces a reduction in stiffness, which, in turn, causes an elongation of the periods of vibration, which then, depending on the shape of the response spectrum being considered (or on the frequency content of an input record), may trigger significant changes in the response characteristics of the buildings The magnitude of this effect is obviously dependent on the structural dynamic properties (i.e. different structural system or different design details) as well as the frequency content of the single record.

As highlighted by Chintanapakdee and Chopra (2003), whilst the level of ductility demand does not affect the base storey drift, the upper storeys drift decreases as ductility increases.

At this stage, it might be opportune to underline the importance of a good estimate of the interstorey drift profile; the latter plays a fundamental role (i) in pointing out when and where higher mode effects

Evaluation of Pushover Procedures for the Seismic Design of Buildings

29

are of relevance, and (ii) in predicting the yield mechanism that the structure will develop (Seneviratna, 1995). Since, interstorey drifts are strongly dependent on both higher mode contributions and distribution (and magnitude) of inelastic deformations among structural members, the quality of their prediction has to be considered an important index of the ability of each pushover scheme to capture these two effects. Moreover, given its direct correlation to both structural and non-structural damage, interstorey drift is one of the most important parameters for damage control, hence its extensive employment in performance based seismic engineering. Therefore, an adequate estimate of the storey drift and its distribution along the height of the building should be reasonably predicted by a pushover procedure.

(i) Current Code Provisions regarding MDOFEs

Current code procedures focus their attention mainly on traditional pushover schemes, where an invariant lateral load pattern is adopted, and usually account for higher mode effects only in a simplified way. Since these constant distribution methods are incapable of capturing aforementioned variations in the characteristics of the structural behaviour under earthquake loading, they usually fail to predict successfully the dynamic response. Current codes accounts for these MDOFE only by means of simplified requirements, as those summarised below:

The Eurocode 8 (prENV 1998-1, 1994) requires that two distinct lateral force patterns have to be considered: storey mass-proportional distribution and a first mode distribution (storey forces proportional to mass and first mode shape).

FEMA 356 (ASCE, 2000) prescribes that two separate pushover analysis have to be done, each using different load vectors. Then, for each response quantity of interest the larger value of the two analyses has to be assumed. Table 2 summarizes the different load vector shapes that have to be considered. In particular, have to be highlighted that the effects of a MDOFS response are related to the magnitude of the mass participation factor in the first mode: simple first mode or codespecified lateral load distribution can be adopted when the first mode has been predicted to dominate the response. Instead, a linear dynamic analysis have to be performed if any SRSS storey shear (including modes representing 90% of the total mass of the system) exceeds 130% of the corresponding storey shear from a first mode response spectrum analysis.

ATC-40 (ATC, 1996) requires that the multi-mode pushover have to be used for structures with period larger than 1 sec.; instead with low period systems the first mode lateral load pattern can be used (or alternatively other load vectors, such as the code distribution, concentrated load, or an adaptive scheme).

Evaluation of Pushover Procedures for the Seismic Design of Buildings

30

Table 2. FEMA 356: lateral load distribution requirements 1st Lateral Load Distribution Code Distribution (mass in first mode>75%) 1st mode (mass in first mode>75%) SRSS (must be used for Te>1sec.) 2nd Lateral Load Distribution

use must

Uniform Adaptive

(ii) Performance of pushover procedures in case of relevant MDOFE

As mentioned in the previous section non-adaptive load vectors can provide only approximate estimates of the dynamic results. The accuracy of these simplified procedures in predicting the different dynamic response quantities (displacement, drift, interstorey shear, and moment) of steel frames, as well as a wall system, of different height and at different deformation levels was the objective of the ATC-55 project. As reported in FEMA 440 (ATC, 2005), results shown that the non-adaptive pushover schemes can lead to:

A reliable estimate of the largest interstorey drift that may occur at any location along the building height, but the correct drift pattern cannot be captured where MDOFE became relevant (flexible high rise buildings).

A poor prediction of interstorey shears and moments, in particular for high rise buildings. These parameters are largely underestimated thus resulting in unconservative results, which reveal the inadequacy of such procedures in the case of this kind of structures.

Thus, considering these results the use of NSPs in the design practice will be limited only to low-rise buildings were the effects of higher modes response is negligible and reliable results can be obtained. Instead, where MDOFEs are of relevance a nonlinear dynamic analysis is recommended. Since variations of dynamic properties of the system and site-specific response amplification cannot be accounted for, NSPs based on a non-adaptive pushover analysis are unable to predict MDOFSs response parameters and thus they can be considered intrinsically limited. In particular, predictions of maximum storey drifts, storey forces, and inelastic deformation demands are not reliable using a single invariant load vector.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

31

In the ATC-55 project is concluded that current procedures using multiple load vectors representative of the fundamental mode and one or more higher modes (Multimode Pushover Analysis (MPA)) could improve these predictions but, since their performance is dependent upon the parameter of interest (drift, forces etc.) and the characteristics of the structure, their applicability have to be verified with future researches. Thus, until now, MPA procedures can be considered only with an appropriate degree of caution for comparison with the results given by a single load vector procedure. They have also found that any single nonlinear response history analysis result often produced better estimates of maximum engineering demand parameters (in particular, interstorey shear and moment) and, according to that conclusion, an alternative nonlinear dynamic procedure is proposed.

(iii) Nonlinear Dynamic Procedure using Scaled Response Histories (Scaled NDP Method) (FEMA 440, 2005)
Based on the studied conducted in the ATC-55 Project an alternative design procedure is suggested. It consists in an analytical procedure in which response quantities are determined by nonlinear dynamic analysis using ground motion records that are scaled so that the peak roof displacement matches a predetermined target displacement. The aim of this method is to use the nonlinear static procedure (in the new updated version) to evaluate the maximum displacement at the roof level (since they can provide enough accuracy) and than use nonlinear response history analyses to investigate the MDOF response (in order to have a better prediction of forces and inelastic deformation demand). The procedure will develop following these steps: 1. Estimate the peak displacement of the control node using a NSP for a given response spectrum representative of the site hazard of interests. 2. Select n ground motions (at least 3) according to the specific site hazard of interests. 3. Scale these records in order to have a control node displacement equal to the one previously estimated. 4. Extract the peak values of the response quantities of interests and compute the sample mean

(x n ) .
5. Account for the sampling error that may exist in the choice of the n records, compute also the response quantities at the mean plus k standard deviation level. A simplified method is proposed to compute these quantities by multiplying the sample mean by a coefficient c that depends on the coefficient of variation of the sample. If the response quantities are normally distributed, the quantity x* exceeds the true mean plus k standard deviations with confidence level . x * = c(1 + kCOV )x n ( 2-6)

Evaluation of Pushover Procedures for the Seismic Design of Buildings

32

c=

1 1 1 ( )

) COV
n

( 2-7)

where:

-1 is the value of the variate of the T-Student distribution with n-1 degrees of freedom and COV is
the coefficient of variation for the sample of n observations of the response quantities x.
2.3.4.5 Innovative strategies

According to the considerations reported above, pushover analyses might be grossly inaccurate for structure of larger periods, where higher mode effects tend to be important, or where large inelastic deformations occur; thus it is fundamental to determine the degree of accuracy in the prediction of the response of such systems using an invariant load pattern as well as the ability of an adaptive scheme to correctly reproduce the dynamic envelope of the inertial forces. The research presented in this document will concentrate on the innovative Displacement-based Adaptive Procedure (DAP) developed by Antoniou and Pinho (2004(b)) and results compared with those provided by traditional pushover schemes. Such adaptive procedure is innovative because:

It accounts for the evolution of damage in the structure through the step-by-step updating of the load vector, which reflects the actual stiffness state of the system. Displacements rather than forces are applied to system in an adaptive fashion, thus allowing for the reproduction of peculiar failure mechanisms. It accounts for higher modes effects through the combination of several modal responses. Specific spectrum scaling is considered in the definition of the forcing vector. It is a single-run procedure, thus easier to implement in a computer code and also easier to adopt for design purposes.

It might be underlined as, before DAP, a force-based alternative (FAP) has been proposed by Antoniou and Pinho, 2004(a). The Force-, as well as the Displacement-, based adaptive pushover approaches have been tested and their accuracy in the prediction of dynamic response of RC frame structures has been evaluated (Antoniou and Pinho, 2004(a)(b); Lpez-Menjivar, 2004). FAP shows improved capacity curves and derives interstorey drift and shear profiles closer to dynamic results at pre-yield deformation stages, over its conventional counterparts. Despite their apparent conceptual superiority, or at least despite their conspicuously more elaborated formulation, the improvement introduced by such Force-based Adaptive Pushover (FAP) procedures was not-

Evaluation of Pushover Procedures for the Seismic Design of Buildings

33

necessarily impressive, with respect to its traditional non-adaptive counterparts, particularly in what concerns the estimation of deformation patterns of buildings, which seemed to be poorly predicted by both types of analysis. As described by Kunnath (2004) and Antoniou and Pinho (2004(a)), the main reason for such underperformance seems to be the quadratic modal combination rules (e.g. SRSS, CQC) used in computing the adaptive updating of the load vector; such rules will inevitably lead to monotonically increasing load vectors, since the possibility of sign change in applied loads at any location is precluded, whilst it may be needed to represent the uneven redistribution of forces after an inelastic mechanism is triggered at some location. With a view to overcome all the limitations described above, Antoniou and Pinho (2004(b)) have proposed a paradigm shift in pushover analysis, by introducing the innovative concept of Displacement-based Adaptive Pushover (DAP), where the floor displacements (i.e. structural response parameters), instead of the applied forces (i.e. actions on the structure) are directly controlled. Contrarily to what happens in non-adaptive pushover, where the application of a constant displacement profile would force a predetermined and possibly inappropriate response mode, thus concealing important structural characteristics and concentrating inelastic mechanisms at a given location, within an adaptive framework, a displacement-based pushover is entirely feasible, since the loading vector is updated at each step of the analysis according to the current dynamic characteristics of the structure. One of the main advantages in using a displacement-based pushover procedure lays on the fact that storey forces or shears are no longer applied directly to the structure but rather come as a result of structural equilibrium to the applied displacement pattern, thus allowing for the reproduction of reversal of storey shear distributions, observed in dynamic analysis, even if a quadratic rule is employed to combine the contribution of the different modes. In effect, DAP drift profiles, despite carrying a permanently positive sign, feature, in any case, changes of their respective gradient (i.e. the trend with which drift values change from one storey to the next) introduced by the contribution of higher modes. When such gradient variations imply a reduction of the drift of a given storey with respect to its adjacent floor levels, then the corresponding applied storey horizontal force must also be reduced, in some cases to the extent of sign inversion, as shown (Fig. 9). In other words, given that in DAP, shear distributions are automatically derived to attain structural equilibrium with the imposed storey drifts, rather than being a result of the loads directly applied to the structure, the previously described limitations evidenced by force-based adaptive schemes that use quadratic modal combination rules can be overcome and, consequently, results as whole (i.e. deformation profiles and capacity curves) become more accurate.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

34

12 11 10 9 8 7 6 5 4 3 2 1 -500 0 500 1000 1500 2000 adaptive uniform triangular

storey shear (kN)

Fig. 9. Storey force distributions of a 12-Storey building obtained with Displacement-based Adaptive Pushover as well with standard non-adaptive pushovers The authors found that DAP manages to provide greatly improved predictions throughout the entire deformation range with respect to the Force-based alternative, even if an exact reproduction of the dynamic response could not be achieved (Antoniou and Pinho, 2004(b)). This can be due by the inherent static nature of this kind of procedures or the limitations of the adopted algorithm, as for example the always additive drift computation. Due to the assessed improvements given by the displacement-controlled procedure (Antoniou and Pinho, 2004(a)(b), Lpez-Menjivar, 2004), in the present work attention is concentrated only over this type of adaptive scheme, that will be further tested and compared with other non-adaptive methods in the case of steel-frame structures.
2.3.5 Adaptive Pushover Algorithm

The Force-Based, as well as the Displacement-Based Pushover procedures have been implemented by Antoniou and Pinho in SeismoStruct (Seismosoft, 2005), a fibre-modeling Finite Element program for seismic analysis of framed structures, which can be freely downloaded from the web. The implementation of the proposed algorithm can be structured in four main stages: (i) definition of nominal load vector and inertia mass, (ii) computation of load factor, (iii) calculation of normalised scaling vector and (iv) update of loading displacement vector. The Displacement- and the Force-Based approaches are essentially equivalent, and they differ only in the steps (iii) and (iv). Since the attention

Evaluation of Pushover Procedures for the Seismic Design of Buildings

35

of the present work will concentrate on the validation of DAP procedures, the algorithm is proposed mainly under the perspective of the Displacement-based approach, referring to the alternative Forcebased method only for sake of completeness. Definition of nominal load vector and inertia mass This first step is carried out only once at the start of the analysis in order to define a nominal load vector U0. Such vector (i) allows for the definition of the structural nodes where the loads (forces or displacements) are applied and (ii) it characterises the initial load distribution shape. The latter is then automatically defined and updated by the solution algorithm at each analysis step, for which reason U0 must always feature a uniform distribution shape, in height, so as not to distort the load vector configuration as determined in stage (iii) at each analysis step, in correspondence to the dynamic response characteristics of the structure. Moreover, the adaptive pushover requires, among the start-up conditions, the inertia mass M of the structure to be modelled, so that eigenvalue analysis performed in stage (iii) may be carried out. Both lumped and distributed mass elements may be employed and freely spread throughout the structure. Computation of load factor The magnitude of the loading vector U at any given analysis step is obtained by the product of its nominal counterpart U0 and the load factor at that step.

U = U 0

( 2-8)

The load factor is automatically increased, by means of a (a) load control or (b) response control incrementation strategy, until a predefined analysis target, or numerical failure, is reached. Whilst in the latter the structural response (e.g. nodal displacements and rotations) is directly incremented and the load factor corresponding to such deformation level can be computed, in the load control strategy the applied load vector is directly controlled. The algorithm (b) is preferred in FAP applications, since it provides direct control of structural deformations (e.g. capture soft-storey mechanisms, model the softening post-peak loading/response range). Instead, in DAP analysis, since the applied loads and response deformations can be considered effectively coincident, the slightly less elaborate load control strategy can be adopted. Steps (i) and (ii) above are essentially the same in DAP and FAP with the only difference that in the former the load vector U consists of displacements whilst in the latter consists of forces (referred as P in Antoniou and Pinho (2004(a)).

Evaluation of Pushover Procedures for the Seismic Design of Buildings

36

Calculation of normalised scaling vector The normalized modal scaling vector D determines the shape of the load vector (or load increment vector), accounting for the actual stiffness state of the structure, at every analysis step. An eigenvalue analysis, according with the Lanczos algorithm (Hughes, 1987), is firstly carried out at the end of the previous load increment, in order to determine the modal shapes and participation factors of a pre-defined number of modes n. This vector is defined by normalizing each floor displacement Di so that the maximum displacement remains proportional to the load factor, as required within a load control framework (as adopted in DAP).

Di =

Di max Di

( 2-9)

The eigenvalue vectors are employed to determine the interstorey drifts ij, whilst the displacement Di at the ith storey is obtained through the summation of the modal-combined interstorey drifts of the storeys below that level.

Di = k
k =1

( 2-10)

with
SRSS ( ij ) i = CQC ( ij )

( 2-11) ( 2-12)

ij = S d , j j (i , j i 1, j )

A spectral amplification factor is also included since the adoption of weighted interstorey drifts, by means of the displacement response spectrum ordinate corresponding to the period of vibration of the jth mode (Sd,j), leads to a significant improvement of the prediction of both capacity curves and drift profiles (Antoniou and Pinho, 2004(b)). Multiple response spectra, derived for varying values of equivalent viscous damping, should ideally be employed so as to reflect the actual energy dissipation characteristics of the structure at each deformation level. The implementation of such refinement, however, is beyond the scope of the current work, for which reason a single constant response spectrum derived for an arbitrarily selected equivalent viscous damping ratio of 5% is used throughout the entirety of each analysis. Moreover, it might also be noticed that such interstorey drift-based scaling method has been preferred against a displacement-based scaling approach, where storey displacement patterns are obtained directly from a combination of the n eigenvalue vectors, because the former allows for an improved prediction of drift profiles.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

37

In the proposed adaptive procedure approximations origin mainly from the modal combination rule that has been assumed. In fact, the CQC (or SRSS) rule can just take into account in an approximate way that all the drift maxima at different storeys do not occur at the same time, and also, due to its always additive nature, negative contributions cannot be considered in the combination. Until a more refined modal combination scheme will be adopted the drift profiles of each vibration mode are combined using either the Square Root of the Sum of the Squares (SRSS, equation ( 2-13)), if the modes can be assumed to be fully uncoupled (i.e. jt=0), or the Complete Quadratic Combination (CQC, equation ( 2-14)) if cross-coupling of modes and respective viscous damping have to be considered.

i = i =

j =1 n

2 ij

( 2-13)

j =1 t =1

ij

jt it
jt 0 jt = tj =1 jj

( 2-14)

jt =

(1 r )

8 2 (1 + r )r 1.5
2 2

+ 4 2 r (1 + r )

( 2-15)

r=

j t

( 2-16)

where n is the number of modes, jt is the cross-modal correlation coefficient between the jth and the tth modes, which can be approximated by the above equation if all modes are assumed to feature the same viscous damping ratio . In all the analyses performed in the present study, a CQC modal combination rule has been adopted, based on the following considerations:

It leads to essentially the same results where modes are uncoupled ( vanished). The correlation coefficient (equation ( 2-15) proposed by Der Kiureghian, 1981) diminishes rapidly as the two natural frequencies move farther apart, especially at low damping ratio, and at any given frequency ratio r it increases as the equivalent damping become larger (Fig. 10).

Even if the combined storey drifts (i) still have only positive values, such scheme may also account for the possibility of sign change in interstorey drifts between different modes. In fact, if the two modal responses have opposite sign the cross-coupling term is negative, thus accounting in a certain way for their different mode shape. It might be noted that, as highlighted by Priestley (2003), the use of CQC (or SRSS) rules to combine modal forces may not always lead to the attainment of an adequate load vector shape that accurately represents the dynamic response

Evaluation of Pushover Procedures for the Seismic Design of Buildings

38

characteristics of the structure at a particular deformation level. In fact, the normalized scaling vector should ideally be obtained through a weighted vectorial addition of the contribution of each mode, in order to have a realistic reproduction of dynamic analysis response envelopes (as discussed in 2.3.4.2). Thus, the CQC rule adopted, even might not be the optimal solution, can be considered satisfactory until the refinement of more accurate direct vectorial schemes. Fig. 10 shows the correlation coefficient vs. the frequency ratio between two modes for different equivalent damping ratios. Can be easily seen that the degree of coupling between the two modes increases with the equivalent damping ratio. The continuous line refers to the 5% equivalent viscous damping, value that has been adopted for both the response spectrum and the CQC rule for all the structural models considered in the present study. This value has been arbitrarily selected considering that the structure will cross several stiffness states during the pushover analysis and thus the amount of the equivalent viscous damping cannot be rigorously defined. The select amount of damping allows also for a smoothing of the response spectrum thus avoiding excessively irregular shapes (produced by lower damping ratios) which may affect in an unrealistic way the scaling of the modal contributions.
1.2 0.02 0.05 0.1 0.2

Correlation Coefficient jt

0.8

0.6

0.4

0.2

0 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2

Frequency Ratio r

Fig. 10. Correlation Coefficient-Frequency Ratio relationship The Force-Based approach differs slightly with respect to DAP procedure. In FAP, storey forces (Fij and Fi) replaces interstorey drifts (ij and i), and the normalized modal scaling vector is obtained by normalizing storey forces with respect to the total value:

Fi =

Fi Fi

( 2-17)

Evaluation of Pushover Procedures for the Seismic Design of Buildings

39

It is noteworthy that when the structural response reaches its post-peak range, the eigen-solver can no longer output real eigen-solutions, due to the presence of negative values in the diagonal of the stiffness matrix which in turn lead to imaginary periods of vibration, corresponding to wholly unfeasible modal shapes. In such cases, with both DAP or FAP procedures, the load vector shape is no longer changed (only its magnitude is updated), effectively meaning that a conventional non-adaptive pushover analysis is employed thereafter. Finally, in cases where a very large number of analysis steps are employed and/or when the structural model is very large, it might prove advantageous for the load vector shape to be updated with a frequency lower than the number of analysis steps, so as to reduce the computational effort. Within the framework of the current study, however, for reasons of accuracy and analysis stability, the load vector shape is updated at every analysis step, with up to ten modes of vibration being considered in its computation. Update of loading displacement vector Once the normalised scaling vector D and load factor t (or load factor increment t) have been determined, and knowing also the value of the initial nominal load vector U0, the loading displacement vector Ut at any given analysis step t can be updated following an incremental updating scheme (equation (2-16)). With incremental updating the load vector, at any given analysis step t, is obtained by adding to the load vector Ut of the previous step Ut-1 a newly derived load vector increment, computed as the product between the current load factor increment t, the current modal scaling vector Dt and the nominal load vector U0.
U t = U t 1 + t D tU 0
normalised shape normalised at step t shape at step t nominal load vector new increment of forces of displacements

( 2-18)

Ptt = t U

=
new increment of displacements of forces

balanced displacements existing forces

new at step t new displacements forces applied applied at step t

U U + P Pt-1 + U Pt= t= t= t-1

Fig. 11. Updating of the loading displacement vector

Evaluation of Pushover Procedures for the Seismic Design of Buildings

40

In incremental updating the current modal scaling vector D t , and thus the load increment at the current step, is calculated according to the actual tangent stiffness state of the structure. The analysis is run in incremental fashion through piecewise linearisations, as schematically shown in Fig. 12, where the tangent stiffness is defined as the secant stiffness at each step increment. In the Force-based Adaptive Pushover scheme the approach is essentially the same, and the loading vector can be obtained by replacing displacement- with the correspondent force-term.
Pt = Pt 1 + t F t P0

( 2-19)

F t+ F t Ft F k+ F k Fk Kk O 'k

O 't

Kt

dk dk+ d k

dt

d t+ d t

Fig. 12. Incremental Updating strategy

3
3.1 Introduction

Case Studies and Analyses Post-Processing

The present study focuses on the verification of the increased accuracy potential of such an innovative displacement-based adaptive pushover method (DAP) to estimate the response characteristics of steel buildings subjected to earthquake excitation. DAP and conventional pushover analyses are therefore carried out and compared with the predictions of inelastic dynamic analysis, in terms of both global and local response. In particular, the attention concentrates on Multiple Degrees Of Freedom (MDOF) effects at different ductility (i.e. total drift) levels. The aim is to provide additional results and thus allow for a better understanding of the performance of static procedures, with respect to two recently proposed research works, one presented in the FEMA 440 (ATC, 2005) and the other one proposed by Goel and Chopra (2004). These two researches have been taken as references for both the structural models and the earthquake records, and two separate parametric studies has been performed following the outline adopted in each reference work. In the following these studies will be mentioned as first and second parametric studies, where the first is based on the work presented in the FEMA 440 and the second has the principle aim to compare results with those obtained by Goel and Chopra (2004).

3.2
3.2.1

Case studies First parametric study

The prototype buildings analyzed in the current endeavour consist therefore of two steel momentresisting frames (three and nine storeys) designed as a part of the FEMA-funded SAC joint venture project (Gupta and Krawinkler, 1999). These frames were considered in both regular and irregular (weak-storey at the ground floor) configurations, thus leading to a total of four frames.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

42

Modified steel frame models, in order to develop a weak storey behaviour at the first level, are also considered, for a total amount of four different structural models. The three and nine storey steel frames prototypes have been designed for Los Angeles using the 1994 Uniform Building code, and employed pre-Northridge special moment-resistant frame connections along the building perimeter (refer to Appendix A and Gupta and Krawinkler (1999) for a detailed description). As defined in UBC (1997) a weak storey develops if the lateral strength is less than 80% of the same in the storey above, and it is associated with a local column sway mechanism resulting in high rotation demand on such columns. In order to develop a weak storey mechanism the strength of the first storey columns is reduced, without changing their stiffness. Thus only the failure mechanism of such models changes, without affecting their elastic properties and thus their periods of vibration (mode shapes are represented in Fig. 20 and dynamic properties of the models are reported in Appendix A). Weak frames have been defined reducing the first storey columns strength to 50% and 60% of their original value respectively for the 3-Storey model and for the 9-Storey model. All the frame systems have been modelled including a gravity column to reproduce the overturning effect of weights acting on the other vertical loads resisting members of the structure (P- effects). The latter has been defined by means of an external column composed by pin-jointed elements, with arbitrary sectional properties, pinned at each storey. In order to render the current results comparable to those obtained in the FEMA-440 research programme, a brief and preliminary verification of the numerical modelling was carried out, considering the outcomes of both modal analysis and standard pushover. As can be observed in Appendix A (Table 11) and in Fig. 15 and Fig. 16, the results obtained in the current work are in very good agreement with those described in FEMA-440, which allowed the study to progress with confidence. A bilinear model has been adopted for steel, defining the post-elastic stiffness ratio (post elastic stiffness/initial stiffness). The post-elastic stiffness ratio was set to 0.01 and 0.008 for the 3-storey and 9-storey frame respectively. A conservative value for the ultimate steel strain of steel frame elements was fixed in 15% (Ballio and Mazzolani, 1999), in order to get meaningful results. Thus dynamic analyses involving higher strain levels and thus the development of a global failure mechanism are neglected from the statistical computation. For the aforementioned reason the 9-storey weak frame system will be analysed, for the ordinary ground motions, at an ultimate drift level of 2.7% (correspondent to a top floor displacement of 1m), instead of the 4% selected in a preliminary stage, since higher values leads to the structural failure of the structure under most of the records, thus affecting a statistical-based evaluation of the results.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

43

Fig. 13. 3-Storey (left) and 9-Storey (right) steel frame structures: plane view

3.2.1.1 3-Storey Frame

It is the North-South lateral force resisting frame of a benchmark building for the SAC project (Fig. 13, left). The building is 36.58m x 54.85m (120ft x 180ft) in plan and 11.89m (39ft) in elevation (Fig. 16), with a 0.61m (2ft) extension from the perimeter column lines to the building edge, and floor-to-floor height of 3.96m (13ft). The bays are each 9.15m (30ft) wide in both directions, with four bays in the NS direction and six bays in the E-W direction. Whilst the frame in the N-S direction consists in four bays, only three of those are effectively resistant to the lateral forces. All connections are momentresistant for these three bays. Instead the forth bay elements are all pin-jointed. The columns are 370MPa steel, and are wide flange for the Moment Resisting Frames (MRFs). The beams are 300MPa steel wide flange sections, and the floor system (a composite concrete and steel construction is used for the floor) is assumed to be rigid in the horizontal plane. Hence, each lateral resisting frame resists one half of the seismic mass associated with the entire structure. The seismic mass is due to various components, including the steel framing, floor slabs, ceiling/flooring, partitions, roofing and a penthouse located on the roof. Considering the whole structure, the seismic mass of the first and second levels is 957t and the third level is 1040t.

Fig. 14. 3-Storey steel frame: vertical view

Evaluation of Pushover Procedures for the Seismic Design of Buildings

44

0.4 0.35 0.3 Base Shear/Weight 0.25 0.2 0.15 0.1 0.05 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Roof Displacement/Height (%)

Fig. 15. 3 Storey steel frame-1st mode pushover SeismoStruct (left), FEMA 440 (right)
3.2.1.2 9-Storey Frame

The building is 45.73m x 45.73m (150ft x 150ft) in plan (Fig. 13, right) and 37.19m (122ft) in elevation (Fig. 17), and floor-to-floor height of 3.65m (18ft) for the first floor and of 3.96m (13ft) for the above floors. The interior bays of the structure contain simple framing, while perimeter frames are momentresisting frames (MRFs). The bays are each 9.15m (30ft) wide, with five bays in both the N-S and E-W directions. Monolithic column pieces are connected every two levels through moment- and upliftresistant connections at 1.83m (6ft) above the center-line of the beam to column joint. Concrete foundation walls and surrounding soil are assumed to restrain the structure at the ground level from horizontal displacement. This lateral restraint has been modelled using joint elements developing only compressive forces. The columns are 370MPa steel, and are wide flange for the MRF. The beams are 270MPa steel wide flange sections. Considering the whole structure, the seismic mass of each level is 1060t and the roof level (9th floor) is 1360t.
0.18 0.16 0.14 Base Shear/Weight 0.12 0.10 0.08 0.06 0.04 0.02 0.00 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Roof Displacement/Height (%)

Fig. 16. 9-Storey steel frame-1st mode pushover SeismoStruct (left), FEMA 440 (right)

Evaluation of Pushover Procedures for the Seismic Design of Buildings

45

Fig. 17. 9-Storey steel frame: vertical view

3.2.2

Second parametric study

Prototype buildings used in the second parametric study are two steel moment-resisting frame buildings designed as a part of the FEMA-funded SAC joint venture project described in ATC-40 (Gupta and Krawinkler, 1999). The two steel frame models under exam refer to the 9- and 20-Storey steel frames and they are designed, according with the Los Angeles local code requirements (ICBO 1994), following pre-Northridge standards. Whilst the 9-Storey frame, since it is the same adopted also in the first parametric study, has been already described, the main properties of the 20-Storey model will be briefly presented in the following section (refer to Appendix A and Gupta and Krawinkler (1999) for a detailed description). A bilinear model has been adopted for steel, defining the post-elastic stiffness ratio (post elastic stiffness/initial stiffness) in order to get the same pushover curve (Fig. 18) for the first mode lateral load distribution reported in Gupta and Krawinkler (1999), which refers to Model M1 in the reference document. In such a way the correct reproduction of both strength and stiffness is ensured. A postelastic stiffness ratio of 0.005 has been fixed for the steel model in the 20-Storey frame. The mode

Evaluation of Pushover Procedures for the Seismic Design of Buildings

46

shapes for the first three modes of vibration are represented in Fig. 21 and dynamic properties of the structure reported in Appendix A. The same limitation on the maximum steel strain used in the first parametric study has been adopted also in this case and dynamic analyses involving higher strain levels and thus the development of a global failure mechanism are neglected from the statistical computation.
3.2.2.1 20-Storey frame

The building is 30.48m x 36.58m (100ft x 120ft) in plan (Fig. 19, right) and 80.77m (265ft) in elevation (Fig. 19, left), and floor-to-floor height of 3.65m (18ft) for the first floor and of 3.96m (13ft) for the above floors. The interior bays of the structure contain simple framing, while perimeter frames are moment-resisting frames (MRFs). The bays are each 6.09m (20ft) wide, with five bays along the N-S direction and six along the E-W direction. Box columns are used at corners and monolithic column pieces are connected every two levels through moment- and uplift-resistant connections at 1.83m (6ft) above the center-line of the beam to column joint. Concrete foundation walls and surrounding soil are assumed to restrain the structure at the ground level from horizontal displacement. The columns are 397MPa steel, and are wide flange for the MRF. The beams are 339MPa steel wide flange sections. Considering the whole structure, the seismic mass of each level is 551t for all floors with the exception of level two and the roof level (20th floor) where the seismic mass is respectively of 564t and 585t.

Fig. 18. 20-Storey steel frame- SAC 1st mode (Model M1) and SeismoStruct 1st mode (red) pushover curves

Evaluation of Pushover Procedures for the Seismic Design of Buildings

47

Fig. 19. 20-Storey steel frame: plan view and elevation (from Ohtori et al., 2003)

Evaluation of Pushover Procedures for the Seismic Design of Buildings

48

9 8 7 6 mode1 mode2 mode3 5 4 3 2 1 mode1 mode2 mode3

Level

-5

-4

-3

-2

0 -1 i

Level

0 1 2 3 4 5 -1.5 -1.0 -0.5 0.0 i 0.5 1.0 1.5

Fig. 20. 3-Storey (left) and 9-Storey (right) frame: mode shapes

20-Storey

mode1 mode2 mode3

20

15

Level

10

0 -1.5 -1.0 -0.5 0.0 i 0.5 1.0 1.5

Fig. 21. 20-Storey frame: mode shapes

Evaluation of Pushover Procedures for the Seismic Design of Buildings

49

3.3

Analytical tool

All the analyses are performed through the finite element analysis program SeismoStruct (Seismosoft, 2005) where two-dimensional models of the analysed structures have been implemented. SeismoStruct is a fiber element-based program for seismic analysis of framed structures, freely downloadable from the internet. The program incorporates both local (beam-column effects) and global (large displacements/rotations effects) sources of geometric nonlinearity as well as the interaction between axial force and transverse deformation of the element. The spread of material inelasticity along the member length is explicitly represented trough the employment of a fibre modelling approach, implicit in the formulation of the inelastic beam-column frame elements adopted in the analysis.
3.3.1

Modeling of frames

Structural members, beam and columns, are modelled using inelastic frame elements based on distributed plasticity-fibre element approach. Instead, joint elements are introduced in order to reproduce pin-joint connections and mono-lateral constraints, as in the case of the 9-storey steel frame. The model takes into account geometrical nonlinearities, associated to both local (beam-column effect) and global (large displacement/rotation effects) response, as well as material inelasticity. A bilinear hysteretic model is adopted for the steel stress-strain relationship of steel beam and column elements. The fibre modeling approach allows for the accurate estimation of the structural damage distribution, the spread of inelasticity across the section area, subdividing the cross section in a sufficient number of fibres, and along the members length, by means of the inelastic cubic formulation suggested by Izzudin (2001). The stress-strain state is obtained through the integration of the nonlinear uniaxial stress-strain response of individual fibres in which the section has been subdivided. If a sufficient number of elements per member are used, the spread of inelasticity along the member length and thus the plastic hinge length of structural members can be accurately estimated. Since energy dissipation through hysteresis is already implicitly included within the nonlinear material modeling and the fibre element formulation, and non-hysteretic type damping was assumed to be negligible within the scope of the present work, no viscous damping was considered in any dynamic analysis in the first parametric study. The choice of neglect any source of additional damping in dynamic analysis takes also into account results of recent researches (Hall, 2006; Priestley and Grant, 2005) that underline the unrealistic nature and the degree of unconservatism implicit within the adoption of a Rayleigh model for the linear damping component. In fact, under certain conditions as for example nonlinear analyses with softening non-linearity, the damping forces generated by the Rayleigh damping matrix might became unrealistically large compared to the restoring forces, resulting in an analysis being unconservative (Hall, 2006). However, a Rayleigh damping model has been adopted

Evaluation of Pushover Procedures for the Seismic Design of Buildings

50

in the second parametric study, following the same approach pursued by Goel and Chopra (2004), in order to render current results comparable to those of the aforementioned research. A 2% equivalent viscous damping ratio has been assigned to the first mode and at the period of 0.2 sec. in the 9-Storey frame, whilst the same damping ratio has been selected for the first and fifth modes in the 20-Storey building. The Newmark scheme (Newmark, 1959) with automatic step adjustment has been adopted for the direct integration of the equations of motion in nonlinear dynamic time history analyses. It is worth mentioning that some degree of approximation is due to neglecting the shear and warping strains across the element section, not yet implemented in the finite element program.
3.3.2

Verification of the structural analysis software for steel frames

The Finite Element package (SeismoStruct) used in the present work for nonlinear analysis, which capability of predict the response of RC structures through the reproduction of pseudo-dynamic tests carried out at the European Laboratory for Structural Assessment (ELSA) of Ispra (Pinto et al., 1996; Guedes, 1997) has been already assessed, is now validated also for the case of steel frame systems. Experimental data on a full-scale steel structural model subjected to quasi-static loading up to serious strength deterioration will be compared with analytical estimate, and thus the ability of the program to simulate the behaviour of steel frames evaluated.
3.3.2.1 Case study

The prototype building is a three-storey, two-span by one-span steel moment frame tested at the University of Kyoto (Nakashima et al., 2006). The test structure plan dimensions are 12 m in the longitudinal direction and 8.5 m in the transversal direction and it has been designed following the most common design considerations exercised in Japan for post-Kobe steel moment frames (Matsumiya et al., 2004 and Nakashima et al., 2006). It might be noted that columns were extended to the approximate mid-height in the third storey, at which level steel braces were connected horizontally to the columns by high strength bolts through gusset plates to allow for the rotation at the column top thus reproducing a rigid diaphragm action at this level. Quasi-static cyclic loading with increasing displacement amplitude was adopted, and either two or three cycles were repeated for each amplitude (Fig. 22). The quasi-static cyclic loading has been applied at miheight of the third storey level through two jacks (one per frame), and the displacement was controlled in terms of the overall drift angle. A load cell attached at the end of each jack measured, by means of a digital displacement transducer that had a resolution of 0.01 mm, the horizontal load applied by the jack. Four strain gauges were glued on each column surface at two cross sections, each located at a distance of 1 m, measured toward the

Evaluation of Pushover Procedures for the Seismic Design of Buildings

51

column mid-height either from the column top or bottom. Since the cross sections remained elastic up to the end of loading with the 1/20 rad amplitude, the bending moments applied at the cross sections were estimated from the correspondent curvatures. The shear force applied to the column was estimated as the sum of the two bending moments divided by the distance between the measured cross sections, and results shown that such estimate were very reasonable (Nakashima et al., 2006). The column axial force was estimated from the average of the strains measured by the eight strain gauges glued to each column. The beam shear force was estimated from the difference between the axial forces exerted into the columns, one located on the top of and the other located underneath the concerned beam.
3.3.2.2 The model

Fig. 23 and Fig. 24 represent elevation and plan view of the test structure. The two lateral resisting frames lie along the longitudinal direction and since they work almost independently only one of them has been modelled.

Fig. 22. Quasi-static loading program

Fig. 23. Elevation of the test structure (unit: mm)

Evaluation of Pushover Procedures for the Seismic Design of Buildings

52

Fig. 24. Plan of the test structure (unit: mm) The columns were made of cold-formed square tubes, beams were made of hot-rolled wide flanges and through diaphragm connection details were adopted. The latter are defined by short brackets shopwelded to the columns and connected horizontally to beams by high strength bolts (Fig. 25, left). Metal deck sheets and wire meshes were placed on the top of beams, and then a concrete slab was placed on site. The column bases, overcoming what it is exercised in the common practice, anchor bolts were fastened in short deep steel beams, which in turn were securely tied down to the strong floor (Fig. 25, right). The full characterisation of member properties is reported in Table 3, and the mechanical characteristics of materials are listed in Table 4.

Fig. 25. Through diaphragm connection (left) and detail of the column base (right) (unit: mm)

Evaluation of Pushover Procedures for the Seismic Design of Buildings

53

Table 3. Member properties (from Nakashima, 2006)

The lateral resisting frame has been modelled through steel beam and column elements and a bilinear hysteretic model according with the measured material properties after testing are adopted. Basic assumptions consist in neglecting panel zone effects, composite actions induced by the concrete slab and in considering a simplified response at the column base, neglecting the slip-type hysteretic response that occur, due to yielding and consequent development of permanent deformations in the anchor bolts. Table 4. Material properties after testing Member Column Section -300x9, BCR295 -300x12, BCR295 -300x16, BCR295 Beam Flange, SN400B Web, SN400B

y (MPa)
512 522 537 375 455

u (MPa)
602 610 609 563 588

3.3.2.3 Results

Results obtained are reported in Fig. 26, in terms of global response (i.e. total base shear vs. total drift), and in Fig. 27 and Fig. 28, where the storey shear vs. storey drift are represented. It might be underlined as the storey shear force was the load applied by the jack in the concerned plane. As we can observe the analytical response provided by the structural engineering software SeismoStruct accurately follows the experimental data, thus assessing the accuracy and reliability of the proposed model. Both strength and stiffness are estimated with a satisfactory level of accuracy for engineering

Evaluation of Pushover Procedures for the Seismic Design of Buildings

54

applications. It might also be noted that the pinching hysteretic response that occurs after large displacement cycles cannot be reproduced because the slip-type of response that was experimentally observed at the column base (due to the accumulation of plastic deformation in the anchor bolts) was not modelled in the current application, given its reduced importance within the scope of the current application.
3.3.2.4 Conclusions

The ability of the finite-element program adopted within the present work to simulate nonlinear dynamic response of steel frame structures to seismic loads has been proven by comparing analytical estimates to the experimental quasi-static tests conducted at the university of Kyoto by prof. Nakashima and his team of researchers (Nakashima, 2006). The numerical model is able to reproduce the cyclic response of the prototype three storey frame under the perspective of the global (total base shear vs. total drift) as well local behaviour (storey shear vs. storey drift). Even if simplifications have been adopted in the definition of the analytical model, both strength and stiffness are satisfactorily predicted by the program, thus assessing the capability of the computer software for structural analysis SeismoStruct to correctly reproduce the response of seismically loaded steel frame structures.

2000 Experimental Analytical 1500 1000 500 kN 0 -0.4 -0.3 -0.2 -0.1 -500 -1000 -1500 -2000 m 0 0.1 0.2 0.3 0.4

Fig. 26. Total base shear vs. total drift

Evaluation of Pushover Procedures for the Seismic Design of Buildings

55

2000

Experimental Analytical

1500

1000

500 kN

0 -0.2 -0.15 -0.1 -0.05 -500 0 0.05 0.1 0.15 0.2

-1000

-1500

-2000 m

Fig. 27. Storey shear vs. storey drift 1st storey

2000

Experimental Analytical

1500

1000

500 kN

0 -0.2 -0.15 -0.1 -0.05 -500 0 0.05 0.1 0.15

-1000

-1500

-2000 m

Fig. 28. Storey shear vs. storey drift 2nd storey

3.4
3.4.1

Ground motions First parametric study

In order to enable a direct comparison with the extensive and most thorough parametric study described in FEMA-440 (ATC 2005), the same structural models and earthquake records that have been used in such work have been adopted here. Hence, considered ground motions include NearField type records (NF records) as well as ordinary records (Ordinary Ground Motions, OGMs). Their main properties are summarized in Table 5, whilst acceleration and displacement response spectra are represented in Fig. 29 and Fig. 30 respectively.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

56

1.80
A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11

1.60 1.40 1.20 1.00

[g]
0.80 0.60 0.40 0.20 0.00 0.0 0.5 1.0 1.5 2.0 2.5

T [s]

Fig. 29. Ordinary Ground Motions: unscaled acceleration response spectrum

0.60
A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11

0.50

0.40

[m]

0.30

0.20

0.10

0.00 0.0 0.5 1.0 1.5 2.0 2.5

T [s]

Fig. 30. Ordinary Ground Motions: unscaled displacement response spectrum OGM (11 records) should represent the range of motions that may be expected at a specific seismic site. They consists of site class C records, approximately considered typical soil conditions, established by selecting strong motion records that do not present near-fault or near-field characteristics (strong velocity pulses, short duration, higher frequency content, etc.). The 11 records selected are those with the largest elastic spectral displacement at a period of 1sec. among ground motions showing magnitude 5.5<Ms<8 and distance to the fault rupture 8<d<20km. The obtained records were generated by

Evaluation of Pushover Procedures for the Seismic Design of Buildings

57

several events, with no event contributing more than 3 records. OGMs (Table 5) differ in terms of amplitude parameters (PGA, PGV, and PGD), frequency content and effective duration (interval between the build up of 5% and 95% of the total Arias Intensity, (Bommer and Martinez-Pereira, 1999). Thus the set of OGM allows for a comprehensive statistical analysis of the earthquake response of the prototype buildings, as described in 3.6. Table 5. 9-Ground Motions characteristics
Identifier EQ Ms Station Component PGA [g] Ordinary Ground Motions A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 Superstitn 11-24-87 Northridge 1-17-94 Loma Prieta 10-18-89 Chi-Chi, Taiwan 8-20-99 Loma Prieta 10-18-89 Northridge 1-17-94 Chi-Chi, Taiwan 8-20-99 Superstitn 11-24-87 Northridge 1-17-94 Imperial Valley 10-15-79 Imperial Valley 10-15-79 Near-Field Ground Motions ERZ LUC RRS SCH Erzican 3-13-92 Landers 6-28-92 Northridge 1-17-94 Northridge 1-17-94 6.7 Sylmar County Hospital Parking Lot 190 0.865 138 5.5 6.7 Rinaldi Receiving Station 213 0.891 186 6.5 7.3 Lucerne Valley Station 280 0.732 147 14 6.9 Erzican Station NA 0.442 126 7 EERL Caltech EERL Caltech EERL Caltech EERL Caltech 6.9 El Centro Array #11 (5058) 230 0.38 42.1 18 USGS 6.9 6.7 6.6 El Centro Imp Co. Cent(01335) Canoga Park-Topanga Can (90053) El Centro Array #2 (5115) 140 0.315 31.5 17 USGS 106 0.356 32.1 16 USC 090 0.258 40.9 27 CDMG 7.6 6.7 Canoga Park-Topanga Can (90053) (CHY101) W 0.353 70.6 32 CWB 196 0.42 60.8 14 USC 7.1 Gilroy Array #3 (47381) 090 0.367 44.7 17 CDMG 7.6 (TCU122) N 0.261 34 35 CWB 7.1 6.7 6.6 El Centro Imp Co. Cent(01335) Canyon Country-W Lost Cany (90057) Gilroy Array #2 (47380) 090 0.322 39.1 14 CDMG 000 0.41 43 11 USC 000 0.358 46.4 23 CDMG PGV [cm/s] Effective Length[s] Source

NF accelerograms, on the other hand, consist of motions recorded close to the epicentre and which contain very strong velocity pulses, originally included in the FEMA-440 work with the objective of verifying, at least in preliminary fashion, the validity of employing pushover methods in areas where

Evaluation of Pushover Procedures for the Seismic Design of Buildings

58

pulse-like near-fault ground motions are likely to occur. For further details the reader is referred to ATC (2005) and Somerville et al. (1997). As highlighted by Iwan (1999), who raises questions about the validity of pushover methods for such pulse-like near-fault ground motions in the case of RC structures, the performance of static analysis procedures need to be further investigated. Perhaps the set of four records and the number of building prototypes, considered in the present work cannot be considered sufficient to obtain exhaustive results. Due to reduced set of NF motions and the site-specificity of their characteristics results will be analysed separately for each single record. The component of NF motion used is oriented in the horizontal direction in which the maximum ground velocity occurs. The total drift level obtained for the four steel frames considered is reported in Table 6. Table 6. Near-Field records: total drift recorded in dynamic analyses Building Model 3S 3Sw 9S 9Sw
3.4.2

ERZ 4.18 3.61 1.68 1.64

LUC 2.09 2.95 1.83 1.93

RRS 5.03 3.79 1.87 2.33

SCH 3.05 2.57 1.78 2.02

Second parametric study

In the second parametric study the two steel frame models under exam are tested for ensemble of 20 ground motions. Such records, adopted for dynamic (and DAP) analyses, refer to the 2/50 set of records of the SAC joint venture (Gupta and Krawinkler, 1999), representing ground motions with a probability of exceedance of 2% in 50 years. Intensity and duration of such records are reported in Table 7. Table 7. SAC joint venture records: PGA and duration
Record LA21 LA22 LA23 LA24 LA25 LA26 LA27 LA28 LA29 LA30 PGA [m/s2] 12.58 9.03 4.1 4.64 8.52 9.25 9.09 13.04 7.93 9.73 Duration [sec] 59.98 59.98 24.99 24.99 14.945 14.945 59.98 59.98 49.98 49.98 Record LA31 LA32 LA33 LA34 LA35 LA36 LA37 LA38 LA39 LA40 PGA [m/s2] 12.71 11.64 7.67 6.68 9.73 10.79 6.98 7.61 4.91 6.13 Duration [sec] 29.99 29.99 29.99 29.99 29.99 29.99 59.98 59.98 59.98 59.98

Evaluation of Pushover Procedures for the Seismic Design of Buildings

59

3.5

The Analyses

In order to evaluate the performances of different pushover procedures in predicting MDOFS dynamic response quantities at different ductility (i.e. drift) levels, results of nonlinear static analyses need to be compared with those obtained in nonlinear dynamic time-history runs. As exposed in Chapter 2, several researchers have tried to develop innovative procedures, adaptive or not, accounting for multi-mode contributions with the aim to overcome these well known limitations of traditional pushover procedures. Among these, the Displacement-based Adaptive Pushover (DAP) developed by Antoniou and Pinho (2004(b)) is the object of this work, and its performance will be evaluated against traditional schemes, in terms of (i) prediction of the pushover curve (capacity curve) of the system and (ii) prediction of the local dynamic structural response quantities. Whilst only the first parametric study is of relevance for what concern the first objective, both the parametric studies will be considered in the evaluation of the accuracy of pushover schemes in predicting local responses.
3.5.1

First parametric study

Under the perspective of the evaluation of the global structural response, pushover curves (capacity curves) for the prototype frames, as provided by each pushover procedure are compared with the envelope obtained from Incremental Dynamic Analyses (IDA, Hamburger et al., 2000; Vamvatsikos and Cornell, 2002; Mwafy and Elnashai, 2000). The latter are defined as a sequence of dynamic analyses with increasing scaling factor in order to get results in terms of total base shear vs. top floor displacement at different levels of ductility demand, as the following figure will graphically represent.
Max Ba se She ar (kN)
120

100

80

60

40

20

0 0 0 .5 1 1.5 2 2.5 3

drift (%)

Fig. 31. Maximum base shear and top floor displacement values obtained with incremental dynamic analyses

Evaluation of Pushover Procedures for the Seismic Design of Buildings

60

Three possible alternatives in the definition of such plot have been taken into account, with the objective to identify which could be the most effective. These three plot configurations are respectively:

Maximum Top floor Displacement vs. Correspondent Base Shear. Correspondent Top floor Displacement vs. Maximum Base Shear. Maximum Top floor Displacement vs. Maximum Base Shear.

A time-window of 0.25sec. around the Maximum value within which to find the Correspondent one is adopted. As stated by Antoniou and Pinho (2004(a)) it has not been possible to unequivocally demonstrate that any of the three available ways of plotting dynamic analysis envelopes is more meaningful than the others, for all structural types and analysis cases. Moreover, Lpez-Menjivar (2004), by comparing them with the pushover curve correspondent to the best pushover procedure obtained for the case studied, identifies that the first way of plotting the IDA envelopes should be the most effective. Since nonseismically designed RC buildings subjected to three different earthquake records have been considered in the work of Lpez-Menjivar (2004), an enlargement to different structural typologies (steel-frames) is needed to get more accurate conclusions, and it will be considered in the present work. Dynamic response quantities of interest are the storey displacements, interstorey drifts, interstorey shears and interstorey overturning moments. The pushover schemes predictions of such parameters are compared with dynamic results following the statistical approach presented in 3.6. The following set of pushover schemes, including both conventional, invariant load pattern, pushover and the adaptive procedure developed by Antoniou and Pinho (2004(b)), is considered:

Uniform static load pattern [Uniform]. Inverted triangular static load pattern [Triangular]. Lateral load distribution according with the first mode shape [1st Mode]. Load pattern according with code distribution [Code]. Displacement-based Adaptive Pushover [DAP], with interstorey-drift-based scaling, incremental updating. CQC (=5%) is defined as modal combination rule.

Whilst a 5% response spectrum is assumed in computing the spectral amplification factors in the aforementioned adaptive procedures, an alternative approach has been tested in the first parametric study in the case of OGMs. This procedure considers spectral amplification according to the average response spectrum of all records at the defined drift level (Displacement-based Adaptive Pushover with

Evaluation of Pushover Procedures for the Seismic Design of Buildings

61

Average Spectrum scaling, [DAP-AS]). Since the NF records involve different drift responses for the same prototype building (Table 6) this alternative approach has not been taken into account. In order to analyse the problem under the perspective of local quantities, the dynamic response at different ductility demand levels has been considered, identified by means of the total drift value (i.e. top floor displacement/building height). Three different drift levels are assumed and each ordinary record has been scaled in order to get the predefined drift level for each prototype building (scaling factors are reported in Appendix B). Response parameters of interest (displacement, drift, shear and moment) at each level location as recorded in time-history analyses are then compared with those predicted by the pushover procedures at the same roof displacement magnitude. Whilst results are compared under a statistical point of view (see 3.6.1) in the case of OGMs, they are analysed for each single earthquake response for NF records. Under the perspective of the global behaviour, true (dynamic) and predicted (pushover) responses are compared in terms of pushover curve, that is the base shear vs. top floor displacement plot. Pushover curves obtained through static analysis schemes are compared with IDA envelopes. The latter, in order to obtain results at different states of deformation, are defined selecting several record scaling factors in order to develop increasing levels of the total drift. In this way IDA results at each single total drift level for the ordinary record could be analysed following a statistical approach analogous to the one adopted for the evaluation of the local response parameters. Moreover, the effectiveness of the three different plot options for IDA envelopes is further assessed, in order to confirm or not the conclusions made by Lpez-Menjivar (2004) and extend them to the case of steel frames.
3.5.2

Second parametric study

In this second parametric study, dynamic analyses are performed for the 9- and 20-Storey LA frames under the selected ensemble of 20 un-scaled records. Whenever failure of the model occurred the correspondent time-histories are neglected in the result post-processing. Dynamic analyses results are compared with results obtained with pushover procedures where the structure is pushed up to the same top floor displacement recorded in each time-history. The same set of static methodologies adopted in the first parametric study has been adopted here, neglecting the Average-Spectrum variant of DAP [DAP-AS], because in this case each pushover analysis involves a different target displacement and thus the procedure does not provide any additional reduction of computational effort compared with DAP. In fact, since a single DAP analysis has to be performed for

Evaluation of Pushover Procedures for the Seismic Design of Buildings

62

each earthquake record in order to get the required target displacement value, the specific ground motion scaling might be directly and rigorously considered without any sort of approximation. Results obtained with rigorous dynamic analyses are compared with those provided by the static procedures in terms of the main structural response quantities of interest (interstorey drifts, interstorey shears and interstorey moments) following the approach presented in the following section.

3.6
3.6.1

Response Statistics First parametric study

Whilst the dynamic response for Near-Field records, due to their specific nature, is analysed for each unscaled record the response for the 11 Ordinary Ground Motions selected represents a set of possible structural responses for such kind of records. In order to account for the inherent randomness with respect to each OGM record, a statistical characterisation of the obtained response quantities is needed. In fact, given the structural model and a statistical population of records, any response parameter, at local (floor displacement, shear, etc.) as well as global (IDA curve) level, represents a random quantity. Therefore, a statistical representation and analysis of response parameters obtained for each OGM record is needed. Involved in this statistical approach are all the analysis procedures influenced by the single record feature, that is, the IDA and DAP analyses (with the exception of DAP-AS, where the average spectrum is employed instead). An approach analogous of that adopted in FEMA 440 (ATC, 2005) has been selected for sake of homogeneity and results comparability. Response parameters, assumed to be Normally distributed, are treated in terms of mean (arithmetic mean, x ), median (central value), and standard deviation, estimated as the square root of the best unbiased estimate of the variance (equation ( 3-1)). Maximum and minimum values are also reported in order to get a more direct feel on the scatter of the data.

(x
N i =1

( 3-1)

N 1

As stated above, this statistical data post-processing approach is adopted for both local as well as global response parameters. In the former case, for each building model, at each drift level and each level location statistical quantities are computed for each response parameters. In the case of global

Evaluation of Pushover Procedures for the Seismic Design of Buildings

63

structural response, where a set of pushover curves/IDA envelopes is given, statistics of the total base shear are computed at each total drift level. In addition, the effectiveness of different static procedures in predicting the local dynamic response is quantified and compared by means of two distinct error measures. The first of these two, E1, the same error used in FEMA 440 (ATC, 2005), provides a direct insight on how inaccurate is the static method (evidently, the mean of a pushover response estimate is computed only in the case of record-dependent DAP analysis, for all other pushover schemes the single response value is used):
E1 = Mean PUSHOVER MeanTH MeanTH

( 3-2)

In the case of OGM response, the mean is the average, among the different accelerograms, of the peak values recorded in each analysis. Obviously, since the conventional procedures lead to deterministic values, only for DAP the mean value for pushover schemes have to be computed. The second error indicator, E2, is defined as the absolute value of the mean pushover response normalized with respect to the mean dynamic response value, thus providing an estimate of the bias of each single pushover scheme. The target optimal value is simply the unity: each static procedure is biased towards underestimating the true dynamic response if the ratio is less than one and overestimating the response if the ratio exceeds one.
E2 = Mean PUSHOVER optimal 1 MeanTH

( 3-3)

Whilst E1 allows for a direct comparison with results obtained in the ATC-55 project, E2 shows directly and in an effective way not only the efficacy of the single pushover procedure but also in which direction (i.e. bias) it will be misleading. Thus, static analysis predictions can be extensively and directly evaluated. As briefly mentioned above, due to the reduced set of NF motions and the site-specificity of their characteristics, results will be analysed separately for each single record. Hence, computation of the E1 and E2 parameters will be carried out considering the maximum individual response obtained for each record, rather than using mean values.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

64

For what concern the pushover curves, the same error quantities are evaluated considering the resultant base shear, from pushover schemes as well as the nonlinear analyses, at each level of top floor displacement obtained in the Time-History (TH) analyses (TH results are considered in terms of Maximum displacement vs. Correspondent base shear, since, as it will discuss later, its seems to be the more meaningful). The same E1 an E2 are calculated and also the maximum and average values of these errors among all the range of displacement amplitude are evaluated. Whilst for the NF records errors are calculated for each record, mean values have been used for the OGMs.
3.6.2

Second parametric study

Results obtained in this study will be presented in terms of median values of the main structural response parameters of interest, interstorey drift, interstorey shear and overturning moment, obtained in dynamic and pushover analyses, together with the bias recorded for static procedures with respect to time-history analyses results. Structural responses predicted by the different pushover procedures have been evaluated at the top floor displacement level obtained in each dynamic analyses and then median values compared. Moreover, the ratio between the estimate provided by the static scheme and the result of time-history run has been considered (i.e. bias), in order to evaluate if the static procedure is overestimating (bias greater than one) or underestimating (bias smaller than one) the dynamic results. The median value of this ratio has been considered, together with its dispersion. The latter is defined as:

=ln(84th percentile)ln(median value)

( 3-4)

4
4.1 Introduction

Capacity Curves

In this chapter the capability of pushover schemes to predict the global structural response in terms of the pushover curve is evaluated. For this purpose only the first parametric study presented in the previous chapter is of relevance and thus will be considered. Results obtained for both the Ordinary Ground Motions (OGMs) and Near-Field ground motions (NF records) will be evaluate in terms of global response of the structure through the comparison between the pushover curves and the Incremental Dynamic Analysis (IDA) results. In order to correctly assess the accuracy of each static analysis algorithm in the prediction of the socalled pushover or capacity curve, one needs first to define the more meaningful way of plotting the IDA base shear vs. top displacement envelopes against which the pushover curves will be compared with. Whilst a comprehensive probabilistic approach will be pursued in the case of the OGMs, results will be analysed separately for each single NF record due to their peculiar properties.

4.2

Ordinary Ground Motions

IDA results, for OGMs, are presented, following a probabilistic approach where the structural response among all the set of possible motions is considered, in terms of average values accounting for their variability through the standard deviation. IDA curves for each record have been first developed and the base shear values obtained at the same level of structural deformation (top floor displacement) have been treated as different outcomes of the random variable Total Base Shear. An analogous approach has been pursed also for the adaptive pushover results since they directly depend upon each record properties. Once the best plot option of the IDA points is established, the performance of the different static procedures will be qualitative (in terms of pushover curve) and quantitative (in terms of error

Evaluation of Pushover Procedures for the Seismic Design of Buildings

66

measurements) evaluated. Errors associated with the pushover schemes are calculated, following the approach presented in Chapter 3, through the comparison, at each displacement level, between the base shear predicted by the static procedure and the one obtained from IDA analyses. Results will be presented in terms of mean and maximum values of E1 among the whole deformation range, and the analysis of the single values of E2 obtained at each level of magnitude of the structural deformation.
4.2.1

IDA curves

Three possible alternatives of IDA results representation have been taken into account with the aim to identify which could be the most effective (extensive results are reported in Appendix C). Indeed three possible alternatives of IDA results representation may be adopted: (1) Maximum Top floor Displacement vs. Corresponding Base Shear, (2) Maximum Top floor Displacement vs. Maximum Base Shear, (3) Corresponding Top floor Displacement vs. Maximum Base Shear (a time-window of 0.25s around a given Maximum value has been used to find its Corresponding counterpart). Previous work by Antoniou and Pinho (2004(a)) and Papanikolaou et al. (2006) indicated that for RC buildings the most effective way of plotting IDA envelopes should be Maximum Displacement vs. Corresponding Base Shear. Herein, the validity in extending such conclusions to the case of steel frames was briefly examined. It was observed that when the response mode of a frame remains essentially unvaried independently of the characteristics and intensity of ground motion to which the building is subjected to, and if no structural post-peak strength degradation occurs, then all three options lead very much comparable results, as can be gathered from the observation of Fig. 32. On the other hand, in those cases where the higher modes of vibration play an important role in the response of the structure, as in the case of the 9-Storey frames considered in this study, which may also enter a post-peak strength degradation response phase, then an unrealistic overestimation of the strength at large displacement levels may be given by options (2) and (3) for plotting the IDA envelopes, as shown in Fig. 32. Additional results with a comparison between the three possible options of plotting and pushover curves are reported in Appendix C. These observations seem therefore to confirm the results obtained in the previous studies mentioned above, for which reason in the present work IDA envelopes were obtained by considering the values of Maximum Displacement vs. Corresponding Base Shear obtained in each dynamic analysis run.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

67

Max. Drift vs. Correspondent Base Shear 8000 6000

Max. Drift vs. Max. Base Shear

Max. Base Shear vs. Correspondent Drift

3S

[kN]

4000 2000 0 0

0.1

0.2 [m]

0.3

0.4

0.5

0.1

0.2 [m]

0.3

0.4

0.5

0.1

0.2 [m]

0.3

0.4

0.5

4000 3000 [kN]

3Sw

2000 1000 0 0

0.1

0.2 [m]

0.3

0.4

0.5

0.1

0.2 [m]

0.3

0.4

0.5

0.1

0.2 [m]

0.3

0.4

0.5

15000

10000

9S

[kN] 5000 0 0

0.5 [m]

1.5

0.5 [m]

1.5

0.5 [m]

1.5

10000 8000

9Sw

[kN]

6000 4000 2000 0 0 0.2 0.4 0.6 [m] 0.8 1 0 0.2 0.4 0.6 [m] 0.8 1 0 0.2 0.4 0.6 [m] 0.8 1

Fig. 32. Alternative representation of IDA results

4.2.2

Qualitative evaluation of the accuracy of the static procedures

From the observation of the pushover curve predict by means of each static procedure and the most realistic IDA envelope previously established (extensive results are reported in Appendix C) the performance of the static methods in predict the structural response under the global perspective can be first evaluated. Results are presented in terms of mean and mean the standard deviation of the Total Base Shear, at every structural drift level, in the case of dynamic results as well as the staticadaptive schemes.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

68

The overall trend of the dynamic results envelope seem to be relatively well captured by all pushover algorithms over the entire range of structural deformation in the case of the 3-Storey steel frames, where the first mode dominates the response, and in the case of the weak-frames, where the failure mechanism of the structure is imposed. In particular, it might be noted as the Uniform scheme performs well against the others, which tend to underestimate the structural strength, in the case of the 3-Storey building (Fig. 33). Instead, it fails to predict the response in the case of the 9-Storey weak frame, where it leads to an overestimate of the initial stiffness and an exaggeration of the softening that will occur beyond the elastic range.
Capacity Curve-3S-OGM
6000 5000
Mean DAP mean+st.dv mean-st.dv

Base Shear [kN]

4000 3000 2000 1000 0 0.00

1st Mode Uniform Triangular Code DAP-AS Max. Drift vs. Corr. Base Shear Mean+St.dev. Mean-St.dev.

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

Top Floor Displacement [m]

Fig. 33. 3-Storey frame

Capacity Curve-9Sw-OGM
8000 7000
Mean DAP

6000

mean+st.dv mean-st.dv 1st Mode Uniform Triangular Code DAP-AS Max. Drift vs. Corr. Base Shear Mean+St.dev. Mean-St.dev.

Base Shear [kN]

5000 4000 3000 2000 1000 0 0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

Top Floor Displacement [m]

Fig. 34. 9-Storey weak frame

Evaluation of Pushover Procedures for the Seismic Design of Buildings

69

In the case of the 9-Storey frames (Fig. 35), where the higher modes effects became of relevance and post-peak strength degradation of the response occurs, DAP shows higher consistency in leading to less erroneous predictions; non-adaptive pushovers would in some cases lead to excellent results (e.g. see Uniform distribution in Fig. 33) and then lead to quite poor pushover curve estimates (e.g. see Uniform distribution in Fig. 35). In detail, the static procedures, with the exception of the Uniform scheme, performs well in the elastic range and they tend to capture the softening behaviour that occurs at large displacement levels, but the tendency is to slightly overestimate the base shear in this range of deformations. Moreover, the Uniform scheme overestimates the elastic stiffness of the structure. In the case of the weak model (Fig. 34) the majority of the pushover schemes provide pushover curve enclosed within possible dynamic outcomes, and in particular DAP provides a pushover curve very close to the mean dynamic curve.

Capacity Curve-9S-OGM
9000 8000 7000
Mean DAP mean+st.dv mean-st.dv 1st Mode Uniform Triangular Code Max. Drift vs. Corr. Base Shear Mean+St.dev. Mean-St.dev.

Base Shear [kN]

6000 5000 4000 3000 2000 1000 0 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60

Top Floor Displacement [m]

Fig. 35. 9-Storey frame

4.2.3

Quantitative evaluation of the accuracy of the static procedures

The accuracy of each static procedure in predicting the dynamic analyses envelope can be quantified in terms of the error parameters E1 and E2, according with what explained in Chapter 3. In this case, the Total Base Shear values at each drift level are considered in the computation of such quantities. As we can observe from the mean and maximum values of E1 for the steel frames (extensive results are reported in Appendix C), the accuracy provided by all the pushover schemes may be considered satisfactory for the 3-Storey frames whilst a larger scatter can be observed in the case of the 9-Storey

Evaluation of Pushover Procedures for the Seismic Design of Buildings

70

models. Average values can be considered adequate since they do not exceed 20% in all the cases, with the exception of the Uniform scheme for the 9-Storey frame, and maximum values range mainly between 20 and 30%. It might be noted as the latter belong from different sources. In fact, in the 3Storey building the maximum errors are associated with wrong prediction of the strength, and thus stiffness, in the elastic range where the base shear values are not so high and even low variations may induce higher error measurements; instead, for the 9-Storey frame, errors are associated mainly to the post elastic behaviour, not well captured by all the procedures. The analysis of E2 confirms these observations with values that are closer to 1 in the inelastic range of structural deformation for the 3Storey building, and that are closer to 1 in the elastic range of response for the 9-Storey frame. In particular, it might be noted as, whilst the Uniform scheme performs better in the case of the 3-Storey frame with respect to the taller one (Fig. 36), all the other procedures improved their predictions with the 9-Storey building (Fig. 37). For the low-rise frame the better predictions are given by the Uniform scheme, whose maximum errors do not exceed 10%. Instead, the Code-specified force distribution leads to the best results, with average E1 below 5% and maximum value around 10% and 12% respectively in the case of the 9-Storey regular and weak frame. In general, it might be observed that:

The accuracy of all the procedures increases from the regular to the weak model. The Uniform distribution should be the more appropriate in the prediction of the pushover curve for the low-rise frames, whose response is dominated by the first mode, and all the other procedures might be considered accurate in the prediction of the 9-Storey frames response, since they provide mean E1 below 10%.

The conclusion that DAP shows higher consistency in leading to less erroneous predictions based on the qualitative observation of results, is corroborated by the overall mean of error measure E2 and its standard deviation (summarized in Table 8), which confirms that DAP is able to provide predictions closer to dynamic values and with a lower scatter. Table 8. Mean error values (E2) obtained for all steel models under the entire set of records

E2
Mean St.Dev.

DAP
1.00 0.07

1st Mode
0.96 0.07

Uniform
1.06 0.10

Triangular
0.96 0.07

Code
0.94 0.07

Evaluation of Pushover Procedures for the Seismic Design of Buildings

71

Errors-E1
100.00% 90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% DAP 1st Mode Uniform Triangular Code 100.00% 90.00% 80.00% 70.00% 60.00%

Errors-E1

Mean Max

50.00% 40.00% 30.00% 20.00% 10.00% 0.00% DAP 1st Mode Uniform Triangular Code

Mean Max

E1

Pushover

E1

Pushover

Fig. 36. 3-Storey frame

Fig. 37. 9-Storey frame

4.3

Near-Field records

The capability of the different static procedures to capture the structural response under the point of view of its pushover curve is evaluated also for the Near Field records. In this case results are analysed separately for each individual record due to their peculiar properties. As done for the OGMs, once the most effective way of represent the dynamic results has been defined, then the accuracy of the pushover procedures is evaluated and compared under the qualitative as well as the quantitative perspective.
4.3.1

IDA curves

Results for the NF records, confirming what has been found for the OGMs, lead to the selection of the Maximum displacement vs. Correspondent base shear as the best option to plot the IDA analyses results (extensive results are reported in Appendix C. An example is reported in Fig. 38 for the 3-Storey frame).
Capacity Curve - RRSMV1
7000

6000

5000 Base Shear [kN]

4000

3000

2000

Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

1000

0 0 0.1 0.2 0.3 Top Floor Displacement [m] 0.4 0.5 0.6

Fig. 38. 3-Storey frame

Evaluation of Pushover Procedures for the Seismic Design of Buildings

72

4.3.2

Qualitative evaluation of the accuracy of the static procedures

From the observation of the pushover curves can be seen as all the procedures performs well in the case of the weak-frames, where the failure mechanism is imposed, whilst a larger scatter between results provided by the different schemes develops for the other models. In particular, it might be noted that:

The Uniform scheme seems to perform well with respect to the others in the case of the 3-Storey frame, where the other procedures tend to underestimate the structural strength for the most of the records (Fig. 39).

Any of the static procedures is able to capture the stiffening behaviour that may develop in some cases for the 9-Storey building (Fig. 40).

Moreover, it has to be noticed as the dynamic response for the 9-Storey frames subjected to the RRSMV1 record, due to the uncertainties related with the definition of the dynamic results envelope in the elastic range, will be neglected in the evaluation of the errors recorded (Fig. 41).

Capacity Curve - LUCMV1

7000

6000

5000 Base Shear [kN]

4000

3000

2000

Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

1000

0 0 0.1 0.2 0.3 Top Floor Displacement [m] 0.4 0.5 0.6

Fig. 39. 3-Storey frame

Evaluation of Pushover Procedures for the Seismic Design of Buildings

73

Capacity Curve - SCHMV1

14000

12000

10000 Base Shear [kN]

8000

6000

4000

Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

2000

0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Top Floor Displacement [m]

Fig. 40. 9-Storey frame

Capacity Curve - RRSMV1
14000

12000

10000 Base Shear [kN]

8000

6000

4000

Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

2000

0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Top Floor Displacement [m]

Fig. 41. 9-Storey frame

4.3.3

Quantitative evaluation of the accuracy of the static procedures

The analysis of the error quantities for the Near-Field record responses confirms what have been concluded looking at the OGMs.

For the 3-Storey steel frame the Uniform scheme should be the more accurate, even if the error values are larger with respect to what have been found for the OGMs, since in this case the mean of E1 is around 12% and the maximum value is close to 36% (Fig. 42). The E1 values are higher

Evaluation of Pushover Procedures for the Seismic Design of Buildings

74

also for all the other procedures and they can reach, as in the case of SCHMV1 record, 20% for the mean and 30% for the maximum value. No significant changes of E2 are recorded at different deformation levels.

In general, the static procedures provide accurate results for the 3-Storey weak frame with average errors that do not exceed 10%. E1 shows that large errors concentrate in the elastic range of deformation.

The static procedures considered in the present study gives almost equivalent results for the 9Storey frame with average and maximum E1 around 20 and 30% respectively (Fig. 43). Instead, the Uniform scheme leads to less accurate results with mean and maximum errors that can exceed 30 and 60%, as in the case of the LUCMV1 record.

The Uniform procedure is the worst also in the prediction of the 9-Storey weak model response with performances similar to those described for the regular frame (Fig. 44). Among the other static schemes the Code-specified force distribution seems to be the most efficient with mean and maximum value of E1 around 10 and 12% respectively. Moreover, DAP gives satisfactory predictions of the total base shear but with large errors (around 15 and 25% for the mean and maximum). It might also be noted that only these two procedures, Code and DAP, do not underestimate in most of the cases the strength of the system at large levels of deformations. In fact, E2 is close to 1 along all the range of structural deformations (Fig. 45).
E1
100.00% 90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% Uniform Triangular 1st Mode Code DAP
100.00% 90.00% 80.00% 70.00% 60.00%

E1

Mean Max

E1

50.00% 40.00% 30.00% 20.00% 10.00% 0.00% Uniform Triangular 1st Mode Code DAP

Mean Max

E1

Pushover

Fig. 42. 3-Storey frame (SCHMV1)

E1
100.00% 90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% Uniform Triangular 1st Mode Code DAP Mean Max 1.8 1.6 1.4 1.2 1

Fig. 43. 9-Storey frame (LUCMV1)

E2

0.8 0.6 0.4 0.2 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

Triangular 1st Mode Code DAP Uniform

Pushover

E2

E1

Top Floor TH displacement [m]

Fig. 44. 9-Storey weak frame (LUCMV1)

Fig. 45. 9-Storey weak frame (LUCMV1)

Evaluation of Pushover Procedures for the Seismic Design of Buildings

75

4.4

Conclusions

Within the present Chapter first the most realistic option of plotting the IDA envelopes has been identified and then the estimates of the total base shear at each level of the top floor displacement have been evaluated and compared. The analysis of results for the Ordinary as well as for the Near-Field records leads essentially to same considerations. The capability of the static procedures to evaluate the global structural response in terms of its pushover curve can be considered satisfactory in the case of buildings where the first mode dominates the response, as in the case of the 3-Storey steel frame, in particular when the failure mechanism is well defined, as for the weak-model. Instead, the predictions became less accurate for the 9-Storey frame at large levels of structural deformations, due to the higher contribution of modal responses different from the first elastic mode associated with the spread of inelastic deformation throughout the structure. In particular, the Uniform lateral load pattern should be preferable in the case of the 3-Storey frame, whilst all the other schemes provide more accurate results in the case of the high-rise frame model, with similar degree of precision. However, any of the static procedures can capture the stiffening behaviour that may develop in the 9-Storey model for some NF records.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

76

5
5.1 Introduction

Storey Response Parameters

In this chapter results obtained for both the first and the second parametric studies are presented with the aim to evaluate the capability of static procedures to estimate dynamic values of the most important structural response parameters. In fact, a pushover method, in order to be fully reliable, must not only reproduce with enough accuracy the trend of the base shear with respect to the reference node displacement, which is required within the adoption of NSPs, but also to accurately estimate values of the force/deformation demand at each storey level. Only in this way the static method might be assessed to be fully reliable and thus adopted as the analysis method in the design of a new structure or in the retrofit of an existing one.

5.2

First parametric study

Results obtained for both the Ordinary Ground Motions (OGMs) and Near-Field ground motions (NF records) in the first parametric study will be evaluated in this section in terms of predictions of the local response parameters at each storey level (floor displacement, interstorey drift, interstorey shear and interstorey moment). After a brief summary of the main findings of the ATC-55 Project (ATC, 2005), where an analogous study has been performed, the main results obtained will be presented. They will be evaluated, in the case of the OGMs, following the statistical approach discussed in Chapter 3. The scatter in the dynamic and DAP results will be evaluated, as well as the accuracy of each static procedure in predicting the average values of the response parameters of interest. Whilst a comprehensive probabilistic approach will be pursued in the case of the OGMs, results will be analysed separately for each single NF record due to their peculiar non-stationary properties.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

78

5.2.1

Overview of FEMA 440 results

Since the recently developed FEMA 440 document (ATC, 2005) represents the starting point of the present research, whose principle aim is to extend these results also accounting for displacement-based adaptive procedures, the most relevant conclusions presented in this document will be briefly summarized here. The same 4 steel building models and set of ground motions adopted in the present study have been used in the development of the ATC-55 Project, where the performance of a set of static pushover procedures have been evaluated. The 1st Mode, Uniform, Code-specified and Triangular force distributions have been considered as well as multimode procedures, such as the SRSS combination of several modal forces and the MPA proposed by Chopra and Goel (2002). The analysis of the multi-degree-of-freedom effects led to the considerations (ATC, 2005) that are briefly summarised in the following sections for the OGMs and NF motions respectively. 5.2.1.1 Ordinary Ground Motions

Due to the wide set of possible ordinary-type records considered, the dispersion obtained in the dynamic response can be first evaluated for each building model at the different deformation levels considered. Results shown that:

The scatter among the dynamic results increases as the contribution of the higher modes became larger, which increases with the level of the inelastic deformation amplitude. The peak floor displacement response shows the smallest variance with respect to what observed for the other response parameters.

The predictions of the storey response quantities provided by the static schemes considered have been compared with the mean value of the dynamic results, and their accuracy evaluated:

In general, the static procedures give a good estimate of the peak floor displacements (with the exception of the Uniform and Code-specified procedures, all the schemes provide mean and maximum E1 that do not exceed respectively 10% and 20%).

Single-mode load vectors fail to predict the interstorey drift of the high-rise steel frame (mean and maximum E1 may exceed respectively 35% and 80%, for the majority of the single-mode schemes). The MPA may improve the estimate of drifts for the 9-Storey frame, but with still high error values (E1 may exceed 20% and 60% for the regular and weak frame respectively). Moreover, the peak value of the interstorey drift recorded over the height of the prototype buildings is well predicted by all the static procedures (with the exception of the Uniform scheme), but its location cannot always be correctly identified.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

79

The static approaches poorly predict interstorey shears and moments. For the regular frames, dynamic interstorey shears often exceeded pushover estimates, at all storeys. For the weak-frames all the procedures (with the exception of MPA) satisfactory capture the first storey value but they strongly underestimate those at the upper storeys. For what concern interstorey moments, whilst the overall pattern is captured by the static procedures, the dynamic values are substantially underestimated. E1 may exceed, for shears and moments, 50% for the 3-Storey frame and 80% for the 9-Storey building.

5.2.1.2

Near-Field Motions

The prediction of the structural response for the NF motions, evaluated separately for each record, is good for all the building models. In particular, the 3-Storey frame response can be accurately predicted. However, the analysis of results for the 9-Storey frame building underlines as the contribution of higher modes mechanisms, when activated, are difficult to be captured by the static procedures. In fact, the development of large interstorey drifts at the upper floor locations cannot be predicted, in particular with a single-mode-based procedure. Moreover, for the 9-Storey weak frame in those cases where the weak storey mechanism does not clearly develop, as for example for the ERZMV1 and SCHMV1 records, all the pushover schemes tends to exaggerate the displacement and drift demand at that storey location. 5.2.1.3 Conclusions

Considering all the set of results might be concluded, as reported in the FEMA 440 (ATC, 2005), that the static procedures can be considered accurate in the prediction of the displacement shape for all the set of buildings, and the drift profile for low-rise structures. Instead, any of these is able to estimate the interstorey shear and moment demand. Moreover, the multimodal procedure considered, that is MPA, seems to produce results that are somewhat more reliable, but the adequacy of these results depends upon the structural response parameter of interest and the characteristics of the structure. For this reason it is suggested only for comparative purposes. The application of static procedures is recommended only for low-rise buildings and for low degrees of inelastic demand. Instead, they provide unrealistic results where the combination of the structural dynamic properties and the design spectrum shape determine an amplification of the higher modes contribution and when the inelastic mode of deformation of the system is well apart from the 1st mode mechanism.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

80

5.2.2

Evaluation of results

The capability of the static procedures to estimate the local response parameters of the 4 building models for the set of the 11 OGMs, at three different structural deformation levels, and for the NF records will be evaluated both in terms of reproduction of the whole pattern along the building height and in terms of error measurement E1 and E2. In particular, results are summarized through the mean and maximum values of E1, as well as their standard deviation, along the different storey level of each prototype building in order to better visualize the accuracy of each scheme in predicting the single storey response quantities. First results for the OGMs will be presented and than responses for the NF records will be evaluated. 5.2.3 Ordinary Ground Motions

According with the statistical approach discussed in Chapter 3, dynamic as well DAP results have been analyzed in terms of their average values accounting for the dispersion of the data through the standard deviation and extreme (maximum and minimum) values. The detailed set of graphical representations of the buildings responses is reported in Appendix D. The accuracy of the static procedures is evaluated by means of the error measurement E1 and E2. Appendix E includes all their graphical representations. Moreover, figures of the mean and maximum values along the building height for each prototype model at all the deformation levels considered are reported in the same Appendix, as well as the representation of their mean and standard deviation in a tabular format. Results for each response parameter will be described accounting for the different structural building model and level of total deformation. First, in order to understand the overall trend of dynamic results as well as adaptive pushover estimates, the dispersion of results for the different building models and degree of inelastic deformation demand has to be evaluated. In Fig. 46 is represented the coefficient of variation (C.O.V., defined as the absolute value of the ratio standard deviation over mean value) for the different building models and for each design response parameter at increasing level of structural deformation (briefly indicated as 1, 2, 3). Results, for what concern dynamic analyses, show that:

The scatter among the dynamic results increases as the contribution of the higher modes increases, and thus higher variability is obtained for the high-rise buildings. Whilst the C.O.V. tend to increase with the deformation level and from the regular to the weak model in the case of displacements and drifts, a different tendency is shown by the interstorey shears and moments. In this case the variability remains fairly constant, or slightly decreases with higher deformation levels, and the weak model does not necessarily leads to a wide scatter of data.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

81

COV-Floor Displacements 0.5 0.4 0.3 0.2 0.1 0 1 2 Level of Deformation 3 3S-Storey 3-Storey weak 9S-Storey 9-Storey weak

COV-Inter-Storey Drifts 0.5 0.4 0.3 0.2 0.1 0 1 2 Level of Deformation 3 3S-Storey 3-Storey weak 9S-Storey 9-Storey weak

COV

(a)
COV-Inter-Storey Shears 0.5 0.4 0.3 0.2 0.1 0 1 2 Level of Deformation 3 3S-Storey 3-Storey weak 9S-Storey 9-Storey weak

COV

(b)
COV-Inter-Storey Moments 0.5 0.4 0.3 0.2 0.1 0 1 2 Level of Deformation 3 3S-Storey 3-Storey weak 9S-Storey 9-Storey weak

COV

(c) (d) Fig. 46. Coefficient of variation of dynamic analyses results: (a) floor displacements, (b) interstorey drifts, (c) interstorey shears, (d) interstorey moments
COV-Floor Displacements 0.3

COV

COV-Inter-Storey Drifts 0.5 0.4

0.2 COV

0.1

3S-Storey 3-Storey weak 9S-Storey 9-Storey weak

COV

0.3 0.2 0.1

3S-Storey 3-Storey weak 9S-Storey 9-Storey weak

0 1 2 Level of Deformation 3

0 1 2 Level of Deformation 3

(a)
COV-Inter-Storey Shears 0.2

(b)
COV-Inter-Storey Moments 0.2

0.1

3S-Storey 3-Storey weak 9S-Storey 9-Storey weak

0.1

3S-Storey 3-Storey weak 9S-Storey 9-Storey weak

COV

0 1 2 Level of Deformation 3

COV

0 1 2 Level of Deformation 3

(c) (d) Fig. 47. Coefficient of variation of DAP analyses results: (a) floor displacements, (b) interstorey drifts, (c) interstorey shears, (d) interstorey moments

Evaluation of Pushover Procedures for the Seismic Design of Buildings

82

The peak interstorey shears and moments show larger variability than the other response parameters in the low and intermediate range of structural deformation.

In the same way the dispersion obtained throughout the whole set of DAP analyses may be evaluated (Fig. 47). In general the coefficient of variation is lower but it follows approximately the same trend of dynamic results. In fact, the variability of data tends to increase with the deformation level and the number of storeys in the steel frame models, in particular for displacements and drifts. Whilst in the case of dynamic analyses the maximum C.O.V. is approximately the same for all the structural response parameters, in the case of DAP the variability of results concentrates on the interstorey drift. Obviously shears and moments have the same C.O.V. since the applied lateral forces could assume only positive values. 5.2.3.1 Storey Displacements

As one can observe from results reported in Appendixes D and E, in general, static procedures predict the floor displacements with an acceptable accuracy, with the displacement-based adaptive pushover scheme leading to slightly superior estimates, in particular in the case of the high-rise steel frames where higher modes contributions are of relevance. Only the Uniform scheme provides wide errors. In particular, E1 may exceed 110% for the Uniform lateral load pattern, as in the case of the 9-Storey weak frame at the intermediate (Fig. 48) and highest deformation levels. Instead, all the other procedures perform very well, with mean values of E1 that in the most of the cases do not exceed 10%. The accuracy of these estimates can be considered adequate in the case of the 3-Storey frame where the maximum values are within 20%, but the errors in predicting the floor displacements are larger for the other models and it may exceed 40% for the 9-Storey weak frame. In particular, for the 9-Storey weak frame, whilst the 1st Mode and Triangular schemes poorly perform in the intermediate range of deformation (by exaggerating the first floor displacement), they perform well in the higher one, where DAP and the Code-specified distribution of lateral forces instead tend to underestimate the first storey response (E2 around 0.57 and 0.78 respectively). If one considers the bias of these estimates along the building height, when the first-mode-dominates the response (3-Storey frames) the accuracy is generally satisfactory, with E2 values of the order of 0.9. For the 9-Storey models, on the other hand, the tendency is for (i) good estimates to be provided for the regular frame, and (ii) overestimation of displacements at the lower storeys (E2 around 1.2) to be obtained for the weak model.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

83

Drift Level: 2% 10 9 8 7 6

Drift Level: 2% 10 9 8 7 6 Storey 5 4 3 2 1 0

Floor

5 4 3 2 1 0

0.2 0.4 0.6 Floor Displacement [m]

0.8

4 6 8 Interstorey Drift [%]

10

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Fig. 48. 9-Storey weak frame

5.2.3.2

Interstorey Drifts

Results reported in Appendixes D and E clearly confirm that, generally speaking, static procedures predict the interstorey drift profiles with less accuracy with respect to floor displacements, as had been shown already in ATC (2005). In addition:

The estimates become worse when higher modes play an important role in the response and as the level of inelastic deformation demand increases; For example, the mean E1 increases from values around 18% at the low deformation level to 50% at the highest deformation level in the case of the 9-Storey weak frame. An opposite trend is shown only in the 3-Storey frame, mainly due to the extremely low values of floor displacements which make results very sensitive.

For the 9-Storey regular frame, whilst most of the static schemes seem to be equivalent in the prediction of drifts at the lowest level of deformation (mean E1 around 18%), DAP performs

Evaluation of Pushover Procedures for the Seismic Design of Buildings

84

better with respect to conventional procedures in the intermediate and high ranges of deformation (with mean E1 around 23%). The Code distribution gives slightly more accurate results (mean E1 within 18%) at the largest level of total drift, where the other schemes provide errors around 30%, and 26% in the case of DAP (

Fig. 49). DAP performs better against non-adaptive schemes for the 9-Storey models, with mean drift estimate that differs of about 30% (maximum error) with respect to the mean dynamic response at intermediate and high levels of deformation, whilst all the other schemes differ by more than 45%. Maximum values are larger for the weak model, where E1 provided by DAP may exceed 50%, and 70% for the other schemes. It might be noted as for this building model the DAP and Code procedures do not overestimate the 1st storey drift as the other procedures do.

The majority of the considered static schemes satisfactory perform in the case of the low-rise buildings. In most of the cases the mean values are within 10%. Higher values are recorded for the 3-Storey weak model where mean E1 may exceed 20%. The Uniform scheme is by far the worst, leading to prediction errors in excess of 70%. In particular, E1 may exceed 70% for the Uniform lateral load pattern, as in the case of the 9-Storey frame at the highest deformation level and the weak model at the intermediate deformation level (

Fig. 49 and Fig. 48) respectively.

Looking at the variability of these estimates along the building height it can be underlined that:

In the case of the 9-Storey frames, DAP provides more accurate results than conventional procedures also in terms of dispersion of estimates along the building height, with standard deviation of E1 that does not exceed 18%.

all non-adaptive static schemes tend to strongly underestimate (E2 value around 0.20) the drift at the upper storeys, a featured that had been already signalled in FEMA-440. DAP, instead, does not seem to be influenced by the storey level of interest, predicting in general very good shapes of the drift profiles, even if under-predicting the magnitude of the latter. This behaviour has been observed in other studies that focused on bridges (Casarotti, 2005) and 3D reinforced concrete buildings (Meireles et al., 2006), which might indicate that the introduction of some sort of dynamic scaling parameter (empirically derived from a series of parametric studies) to correct the prediction of the interstorey drift profile might be all that is needed to render DAP results essentially coincident with dynamic analysis observations.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

85

Drift Level: 4% 10 9 8 7 6 5 4 3 2 1 0
Storey 10 9 8 7 6 5 4 3 2 1 0

Drift Level: 4%

Floor

0.5 1 1.5 Floor Displacement [m]

5 10 Interstorey Drift [%]

15

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Fig. 49. 9-Storey frame

5.2.3.3

Interstorey Shears and Moments

As will be subsequently shown, the percentage errors observed in the predictions of storey shears and moments are higher that those associated to displacements and interstorey drifts. The overall trend is for underestimation to occur at the upper storeys, particularly when the structure is pushed far into the inelastic deformation range. Prediction errors (E1) associated with these two response parameters exceed 40% on all models, with the advantage of DAP residing on the fact that a narrow scatter in its results, and thus a reduction of its maximum error, is observed. In detail, the mean errors associated with the storey shears and moments are of the order of 25-30% for the frame structures analysed. Maximum values are of the order of 40-50% in the case of the 3Storey models, and increase up to 60-70% for the 9-Storey frames.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

86

As we can observe from values of E2 reported in the Appendix E the overall trend is to underestimate dynamic shear and moment demand, in particular at the upper storey locations. In fact, as reported for example in Fig. 50 for the 9-Storey frame at 4% total drift level, the interstorey shears (and moments follow the same trend) may be strongly underestimated (E2 may be less than 0.2). Moreover, in general, the dispersion of these estimates along the building height is, in most of the cases, larger than what provided for drifts and displacements. Only in the case of the 9-Storey models the variability in the prediction of shears and moments is of the same order of magnitude of the drift estimate. However, it might be noted as DAP leads in most of the cases to estimates closer to dynamic results with a narrow scatter along the building height.

E2 - Inter-storey Shear 1st storey 7th storey 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP - AS

Fig. 50. 9-Storey frame-drift level 4%: Interstorey shear (E2) Considering that the static procedures (with the exception of Uniform scheme) provide similar results for buildings whose dynamic response is dominated by the first mode, as in the case of the 3-Storey models, DAP could finally be considered superior in capturing the MDOFS response, because:

Performances of the different pushover schemes are similar, and without the required accuracy, in the prediction of shears and moments. In general, the displacement profile is satisfactory captured by the static methods. In general, all the static procedures provide good estimates of the structural parameters when the first mode dominates the response, as in the case of the 3-Storey models.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

87

DAP provides a better prediction of the interstorey drift, with a narrow scatter, for the high-rise buildings (i.e. 9-Storey models).

Finally, it can be underlined as the predictions of all the parameters obtained with the Displacement based Adaptive Procedure where the average spectrum has been considered for scaling (DAP-AS) are in all similar with those provided by the mean value among the DAP analyses. Thus, DAP-AS might be used as a simplified procedure in the evaluation of the structural response over a set of possible records, alternative to a statistical evaluation of the DAP results. This represents a significant improvement in the application of nonlinear static procedures in the common practice because it allows, with a single run analysis, a more accurate prediction, if compared with conventional methods, of the average dynamic results within an ensemble of design ground motions. 5.2.4 Near-Field Motions

Results for the NF records have been analyzed considering the building response for each single accelerogram due to the extreme variability in the non-stationary properties of each ground motion. The detailed set of graphical representations of the buildings responses for the 4 NF motions, and the correspondent error measurement, are reported in Appendixes D and F respectively. Within this section the accuracy that the static procedures may provide in predict each one of the response parameters of interest, will be evaluated and results summarized in terms of mean and maximum values of the error quantity E1 obtained among the different records, in order to make the comparison easier. 5.2.4.1 Storey Displacements

As we can observe from results reported in Appendixes D and F, in general the static schemes satisfactory predict the floor displacements for the whole set of building models and records. Mean values of E1 are within 10% for the 3-Storey models, and within 15% in the case of the high-rise frames. Moreover, it has to be noted as the Uniform scheme provides less accurate results with respect to the other pushover methods, particularly for the 9-Storey weak frame where the mean error associated with this procedure exceed 30%. The static procedures, with the exception of the Uniform scheme, which leads to wider errors, provide maximum values of E1 of the order of 20-30% for the 3-Storey weak and regular model respectively. In the case of the 9-Storey regular frame (Fig. 51 and Fig. 52) the 1st Mode and the Triangular schemes seem to perform better (maximum E1 around 25%, whilst it exceeds 40% for the other schemes). Instead, the DAP and Code-specified procedures provide better estimates of the floor displacements in

Evaluation of Pushover Procedures for the Seismic Design of Buildings

88

the case of the 9-Storey weak frame (maximum E1 around 40-45%, whilst it exceeds 70% for the other procedures). From the observation of the error parameter E2, it can be observed as the accuracy in the prediction of the floor displacements is fairly independent with respect to the storey level. Only the Uniform scheme clearly emphasizes displacements at the lower storeys. 5.2.4.2 Interstorey Drifts

As we have observed in the case of the floor displacements, all the static schemes satisfactory predict the interstorey drifts for those building models where the first mode dominates the response. Mean values of E1 are within 10% for the 3-Storey regular model, and within 15% in the case of the 3-Storey weak frame. Moreover, it has to be noted as the Uniform scheme results less accurate than the others, particularly for the 3-Storey weak frame where the mean error associated with this procedure exceeds 35%. The static procedures, with the exception of the Uniform, which leads to wider errors, provide maximum values of E1 of the order of 30-40% for the 3-Storey regular and weak model respectively. Predictions are less accurate for the high-rise models. Mean E1 is around 25 and 35% for the regular and weak model respectively for the majority of the procedures, whilst the Uniform gives higher errors, around 45% and 65% respectively. In the case of the 9-Storey frames DAP performs better than the other methods. In fact it provides maximum E1 around 55-65% for the regular and weak model respectively, whilst it exceeds 70% and 80% for the other procedures. From the observation of the error parameter E2, it can be seen as the accuracy in the prediction of the interstorey drifts depends on the storey level of interest. Considering the whole set of models, and particularly the 9-Storey frames (as represented in Fig. 53 for the regular model) where this variability is emphasized, DAP provides more accurate results since the interstorey drifts are predicted with a good accuracy along the whole building height. Instead, the other procedures, in general, teand to overestimate drifts at the lower storeys, and seem unable to capture higher modes responses which are associated with a large drift demand at the upper floors.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

89

10 9 8 7 6

10 9 8 7 6 Storey 5 4 3 2 1 0

Floor

5 4 3 2 1 0 0.1

0.2 0.3 0.4 0.5 0.6 Floor Displacement [m]

2 3 Interstorey Drift [%]

10 9 8 7 6 Storey 5 4 3 2 1 0 Storey 2000 4000 6000 8000 Interstorey Shear [kN] 10000

10 9 8 7 6 5 4 3 2 1 0

0.5 1 1.5 Interstorey Moment [kNm]

2 x 10
5

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 51. 9-Storey frame-RRSMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

90

10 9 8 7 6 5 4 3 2 1 0 0.1 Storey 0.2 0.3 0.4 0.5 Floor Displacement [m] 0.6 Floor

10 9 8 7 6 5 4 3 2 1 0 0.5

1.5 2 2.5 Interstorey Drift [%]

10 9 8 7 6 Storey 5 4 3 2 1 0
Storey

10 9 8 7 6 5 4 3 2 1 0

2000

4000 6000 8000 Interstorey Shear [kN]

0.5 1 1.5 Interstorey Moment [kNm]

2 x 10
5

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 52. 9-Storey frame-SCHMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

91

E2 - Inter-storey Drift 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

Fig. 53. 9-Storey frame-SCHMV1: Interstorey drift (E2)

5.2.4.3

Inter- storey Shears and Moments

In general, the estimates of shears and moments lead to similar errors and they are worst than those of the other response parameters. Mean values of E1 are of the order of 20-25% for the frame models. The Uniform scheme gives less accurate results with respect to the other pushover schemes, particularly for the 3-Storey weak frame, where the mean error associated with this procedure exceeds 30% in the case of shears and 35% in the case of moments, and in the case of the 9-Storey regular model where the mean error in the prediction of the storey moments may be around 55%. For this high-rise building model DAP performs better against the others in the prediction of the storey moments, with mean error around 10%. It might be noted as the error associated with these response parameters may reach very large values. For the 3-Storey models maximum values of E1 are of the order of 55-65% in the case of the Uniform procedure, and around 40% for the other schemes. Maximum values are larger in the case of the 9-Storey models. In detail, the majority of the static schemes lead to maximum errors greater than 60%. Among these, it has to be underlined as the Uniform lateral load pattern performs badly for the regular frame (with maximum E1 higher than 160%), whilst DAP provides better results, in comparison with the other pushover methods, with maximum errors around 45%.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

92

From the observation of the other error parameter considered, E2, it can be observed that the accuracy in the prediction of the interstorey shears and moments depends on the storey level of interest. Considering the whole set of models, and particularly the 9-Storey frames, where this variability is emphasized, DAP provides more accurate results because the interstorey drifts are predicted with a better accuracy along the whole building height, if compared with the other schemes (an example is reported in Fig. 54 for the 9-Storey frame). It might be noted that, in general, the pushover procedures tend mainly to underestimate drifts at the upper storeys, whilst the Uniform scheme shows the opposite trend.
E2 - Inter-storey Shear 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

Fig. 54. 9-Storey frame-RRSMV1: Interstorey shear (E2)

5.2.5

Conclusions

Within this first parametric study the accuracy of the different static pushover methodologies in the prediction of the dynamic response of the 4 prototype buildings has been evaluated for a set of Ordinary-type- as well as Near-Field-type-records. Considering both groups of records, results seem to confirm what has been obtained in the ATC-55 project. In particular, the capacity of static procedures to predict the structural response in terms of floor displacements, and their limitation in the estimate of the other response quantities, in particular for high-rise buildings and large deformation levels. Results also confirm that the Uniform lateral load pattern leads in the most of the cases to unrealistic results. Moreover, the current study shows that, even it cannot provide the optimal solution, the displacementbased adaptive scheme could represent an improvement with respect to the other static procedures. In

Evaluation of Pushover Procedures for the Seismic Design of Buildings

93

fact, DAP results superior because it provides, with a lower variability along the building height, better estimates of drift, shears and moments, in particular for the high-rise buildings where the single mode schemes are fundamentally limited. Instead, the static procedures provide similar results in terms of floor displacement, and in general for low-rise models. In Fig. 55 results in terms of mean and standard deviation of E1 along the building height obtained for the OGMs at the intermediate deformation level (DL2) are represented as an illustrative example.
Floor Displacement 100 3S - DL2 - E1 - [%] 75 50 25 0 100 3Sw - DL2 - E1 - [%] 75 50 25 0 100 9S - DL2 - E1 - [%] 75 50 25 0 Mean St.Dev. Interstorey Drift Interstorey Shear Interstorey Moment

100

R 9Sw - DL2 - E1 - [%]

75 50 25 0
Code Code DAP Code DAP DAP-AS Uniform Uniform Uniform 1st Mode 1st Mode Triangular 1st Mode 1st Mode Uniform DAP-AS DAP-AS DAP-AS Triangular Triangular Triangular Code DAP DAP

Fig. 55. Mean and standard deviation along the height of the buildings for the whole set of models at the intermediate deformation level (E1) The dispersion of dynamic analyses results has been also evaluated. Since it tends to increase with the magnitude of the deformation demand, in particular for what concern floor displacements and drifts, and it is large for storey shears and moments also at low levels of structural deformation for high-rise

Evaluation of Pushover Procedures for the Seismic Design of Buildings

94

buildings, an adequate prediction of the response in these cases is more difficult and it partially explains the less accuracy of the estimates provided by the static methods. Results obtained for the NF records in general confirm those obtained for ordinary-type motions, even if the level of accuracy might be lower (both mean and maximum error increases). In Fig. 56 the mean and maximum errors obtained throughout the whole set of NF records and building models are summarised in order to provide a synthetic representation of results. However, should be underlined that, since results have been obtained by averaging values among different structural models and different earthquake motions involving different drift demands, a rigorous probabilistic interpretation of such results cannot be performed.
Floor Displacement 50 NF - E1 - [%] Interstorey Drift Interstorey Shear Interstorey Moment

25

0 DAP 1st Mode Uniform Triangular DAP 1st Mode DAP 1st Mode DAP Code

Triangular Code

Uniform Triangular

1st Mode Uniform

Uniform

Code

Fig. 56. Mean values among the whole set of models for all the NF records (E1)

5.3

Second parametric study

In this section results obtained in the second parametric study will be presented. After a brief overview of the main findings obtained in the reference research work (Goel and Chopra, 2004), results obtained will be presented under the perspective of the prediction of the major storey response parameters along the height of the buildings. 5.3.1 Overview of MPA results

The research performed by Goel and Chopra (2004) has been taken as a reference work for the second parametric study conducted in this research. The authors analysed the performance of static procedures, including the Multimode Pushover Analysis (MPA) that they proposed, for six different steel frame models. Prototype buildings were a 9- and a 20-Storey frames, each considered in three variants. They consist in different design of the same structure according respectively with the Seattle (ICBO, 1994), Boston (BOCA, 1993) and Los Angeles (ICBO, 1994) local building codes.

Triangular

Code

Evaluation of Pushover Procedures for the Seismic Design of Buildings

95

Fig. 57. Interstorey drift profile (from Goel and Chopra, 2004)

Fig. 58. Interstorey drift ratio (from Goel and Chopra, 2004) Results, analysed under the perspective of the interstorey drifts and plastic rotations estimates, show that all the static approaches, and MPA in particular; perform better in the case of Seattle and Boston

Evaluation of Pushover Procedures for the Seismic Design of Buildings

96

frames, where the response under the selected input records is nearly elastic, whilst the accuracy decreases as the structure respond in its inelastic regime, as in the case of LA frames. In fact, as one can observe in Fig. 57 and Fig. 58 where the interstorey drift profile and the interstorey drift ratio (i.e. bias, the ratio between the pushover estimate and the dynamic value) are represented, the MPA developed and proposed by the authors performs well in the case of Seattle and Boston frames, whilst results became worst in the case of LA frames. In particular, MPA improves results, with respect to traditional pushover schemes, for buildings that respond in the elastic range but overestimate the drift demand at lower storeys in the 9-Storey LA model and at upper storeys in the 20-Storey LA frame. Moreover, as stated by the authors, the bias in MPA is unacceptably large for the 20-Storey LA building that experience roof drift in the region of rapid decay in lateral capacity of the building, and thus MPA should be abandoned for more rigorous nonlinear dynamic analyses. Such underperformance of the MPA is even more clear accounting for the fact that, in order to avoid the reversal of direction of the roof displacement in the second mode pushover analysis of the 20-Storey LA frame soon after initiation of yielding as the intensity of the lateral load is increased, only partial gravity load was included in the analyses. Similar results have been obtained for an improved version of the procedure (Goel and Chopra, 2006) specially geared towards the calculation of internal forces. In this work, therefore, high-rise steel frames responding in the inelastic range of deformations that have been used by Goel and Chopra (2004) to demonstrate the inability of current nonlinear static procedures in predicting their response, will be analysed using the recently proposed Displacementbased Adaptive Pushover algorithm (Antoniou and Pinho, 2004(b)) with a view to assess the capability of the latter in overcoming some of the shortcomings discussed above. 5.3.2 Evaluation of results

The capability of the static procedures to estimate the local response parameters of the 2 building models for the set of 20 ground motions considered will be evaluated both in terms of reproduction of the whole pattern along the building height and in terms of the dispersion in the estimate provided by the static methodologies, according with the statistical approach discussed in Chapter 3. Attention is focused in particular on the accuracy that pushover procedures might provide in the case of high-rise steel frames. In the following sections will be presented results obtained in a preliminary set of analyses and other results obtained after an empirical calibration of the adaptive pushover procedure. 5.3.3 Preliminary study

Evaluation of Pushover Procedures for the Seismic Design of Buildings

97

In the preliminary stage of the analysis the performance of DAP is compared with predictions provided by non-adaptive procedures for the high-rise steel frames under exam (the full set of results is reported in Appendix G). Results shown that, even if estimates are not completely satisfactory, in particular for the 20-Storey frame, DAP predicts the shape of the drift profile with superior accuracy, in comparison to conventional procedures (and also with respect to what is observed in Fig. 57 and Fig. 58), and thus it represents an alternative simpler procedure (involving a single pushover analysis) that allows to predict the response shape of tall steel buildings with very good accuracy, at least equal with respect to more complex procedures, such as the MPA (Chopra and Goel, 2002). In fact, it estimates the shape of the interstorey drift profile (in Fig. 59 results are represented in terms of bias) much better with respect to non-adaptive schemes, in particular for what concern the response at the upper storeys, with a level of accuracy similar for example to the one provided by MPA. This characteristic of DAP is particularly positive and encouraging, since interstorey drift profiles provide valuable information on the yield/failure mechanism that the structure will develop and can also be directly related to structural and non-structural damage. Moreover, it is evident that the displacement-based adaptive methodology preserve its reliability also in the case of structural models, as for example the 20-Storey LA frame, that experience roof drifts in the region of rapid decay of lateral capacity of the building, where MPA cannot be expected to provide satisfactory estimates of seismic demands and should be abandoned. In such case, DAP fails to predict interstorey drifts no more than 50%, whilst MPA leads to a higher bias at the upper storeys. It might also be noted that DAP is insensitive to peculiar structural responses that might induce reversal in the higher modes pushover curve, as for example occur in the MPA, where particular countermeasures, such as the partial application of the total gravity loads (as for example in the case of the 20-Storey frame under exam (Goel and Chopra, 2004)) or the assumption of a linear structural response in the higher modes (Goel and Chopra, 2006), have to be accounted for. The superiority of DAP with respect to non-adaptive schemes is confirmed also in terms of prediction of internal forces (interstorey shear and overturning moments). Compared with the other pushover procedures, which strongly underestimate values at the upper storeys, DAP leads to results closer to dynamic values, as clearly represented in Fig. 60 where results regarding interstorey overturning moments are represented. From the observation of the dispersion of the bias along the height of the prototype structures (see Appendix G), we can observe that the scatter of DAP values is of the same

Evaluation of Pushover Procedures for the Seismic Design of Buildings

98

order of magnitude of the other procedures, and it is satisfactorily low, implying that results fall within a narrow range of values, thus confirming the reliability of the adaptive pushover estimates.
9S LA 20 9 18 8 16 7 6 Storey 5 4 3 2 1 Storey 14 12 10 8 6 4 2 0 0.5 1 1.5 Interstorey Drift 2 0 0.5 1 1.5 Interstorey Drift 2 20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

Fig. 59. Interstorey drift: bias for the 9-Storey (left) and 20-Storey (right) frames

9S LA 20 9 18 8 16 7 6 Storey 5 4 3 2 1 Storey 14 12 10 8 6 4 2 0 0.5 1 1.5 Interstorey Moment 2 0

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

0.5 1 1.5 Interstorey Moment

Fig. 60. Interstorey overturning moment: bias for the 9-Storey (left) and 20-Storey (right) frames

Evaluation of Pushover Procedures for the Seismic Design of Buildings

99

5.3.4

Extensive study

DAP, as underlined in the previous section, improves results against non-adaptive procedures. Although the actual values of drift are slightly underestimated by the displacement-based adaptive pushover scheme (DAP), the latter predicts the shape of the drift profiles with superior accuracy in comparison to conventional pushover. Hence, whilst the under-prediction of the actual values can be overcome with relative ease by the introduction of empirically-derived factors, the correct prediction of the interstorey drift profile requires an algorithm that is able to take due account of both higher mode contributions and distribution of inelastic deformations among structural members. Such conclusion suggested the proposal of two alternative modified DAP procedures based on empirical observations. The first, account for the bias in the interstorey drift estimates by mean of a new target displacement in the analysis (named, Displacement correction factor method). In the second, instead, results obtained in DAP analyses are directly corrected through the adoption of a proper scaling factor (named, Scaling factor method). These two approaches, since further parametric/statistical studies considering a wider set of building models and ground motions are required in order to derive the empirical parameters, have only the objective to provide suggestions for future research studies. 5.3.4.1 Displacement correction factor method

The first proposal in order to improve DAP results consists in the adoption of a modified target displacement in the pushover analysis. In fact, from the observation of the bias in interstorey drifts (Fig. 59) it might be noted that results are almost constant along the height of the building. Such almost constant pattern suggested that results might be improved with the introduction of a displacement correction factor. The basic idea is that, assuming that displacements continuous to increase with the same rate along the height of the building starting from the displacement profile obtained in preliminary analyses, an higher target displacement might be assumed in order to more closely estimate the dynamic interstorey drift pattern. The aim is to push the structure, assuming that the shape of the interstorey drift profile remains fairly constant, up to a target displacement larger than the top floor displacement recorded in the dynamic analysis (as adopted in preliminary analyses) in order to estimate the true dynamic interstorey drift profile. Since the bias obtained in this preliminary set of analyses is fairly constant along the height of the building models, the target displacement used in DAP analyses, under the assumption made above, might be the dynamic value of the top floor displacement divided by the value of the bias recorded in preliminary analyses.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

100

1 0.9 0.8 0.7 Drift ratio . 0.6 0.5 0.4 0.3 0.2 0.1 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Number of storeys

Fig. 61. Bias in interstorey drift: median values along the height of the structure for the whole set of structural models adopted in both the first and the second parametric studies and intermediate value adopted (red line) In order to estimate such correction factor, the whole set of building models, including those adopted in the first parametric study, has been considered. In Fig. 61 the median value along the height of the building of the bias in interstorey drift is reported for all the models analysed. As one can observe, such ratio is dependent upon the height (i.e. number of storeys) of the structure, and it ranges from about 0.9 to 0.6. In order to have only one value of the correction factor that might be adopted for buildings of different height, an intermediate value of 0.75 will be adopted in order to define the correction factor in DAP analyses. Thus, such analyses will be repeated adopting the value of the target displacement used in the preliminary set of analyses amplified by the factor 1/0.75. Results obtained with DAP, where such a displacement correction factor has been adopted, are compared with those provided by traditional pushover schemes and values from dynamic time-history analyses. Performances are evaluated in terms of interstorey drifts and internal forces, interstorey shears and overturning moments. As we can observe in Fig. 62 and Fig. 63 the improved version of DAP leads to an estimate of the interstorey drift profile closer to dynamic values. In particular, the bias is close to one all over the height of the 9-Storey frame and at the lower storeys in the 20-Storey model (Fig. 62 and Fig. 63). Although DAP captures the overall trend of the drift profile, interstorey drifts tend to be underestimated at the upper storeys of the 20-Storey frame. However, also in this case, results have been improved with respect to those obtained in preliminary analyses, with a bias closer to one.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

101

9S LA 20 9 18 8 16 7 6 Storey 5 4 3 2 1 Storey 14 12 10 8 6 4 2 0 2 4 6 Interstorey Drift [%] 8 0

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

2 4 Interstorey Drift [%]

Fig. 62. Interstorey drift profile: 9-Storey (left) and 20-Storey (right) frames

9S LA 20 9 18 8 16 7 6 Storey 5 4 3 2 1 Storey 14 12 10 8 6 4 2 0 0.5 1 1.5 Interstorey Drift 2 0 0.5

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

1 1.5 Interstorey Drift

Fig. 63. Interstorey drift: bias for the 9-Storey (left) and 20-Storey (right) frames The improved adaptive methodology provides superior results also in the prediction of internal forces. As represented in Fig. 64 where the interstorey overturning moment profile is represented as an illustrative example (the full set of results is reported in Appendix G), DAP leads to predictions very close to dynamic values.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

102

9S LA 20 9 18 8 16 7 6 Storey 5 4 3 2 1 Storey 14 12 10 8 6 4 2 0 1 2 Interstorey Moment [kNm] 3 x 10

5

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

1 2 3 Interstorey Moment [kNm]

4 x 10
5

Fig. 64. Interstorey overturning moment profile: 9-Storey (left) and 20-Storey (right) frames Finally, the dispersion of the ratio between static and dynamic analyses (results are reported in Appendix G) has been evaluated. Results are similar with those obtained in preliminary analyses with DAP providing fairly low values, of the same order of magnitude of those recorded for the other conventional procedures. 5.3.4.2 Scaling factor method

As briefly mentioned before, in order to overcome the drift under-prediction of DAP, which, by the way, has been observed also in other studies for different structural typologies (e.g. Casarotti, 2005; Meireles et al., 2006), one could perhaps think of deriving, though a series of parametric studies, empirical scaling parameters in order to account, even if in a simplified way, for dynamic effects. The second approach proposed consists in correcting the interstorey drifts provided by DAP accounting for the bias obtained in the preliminary study. For example, one could consider for instance to divide the static response previously obtained by the correspondent bias (which provides a measure of how much dynamic results are underestimated/ overestimated), averaged throughout the height of the structure. Application of such correction factors (Table 9) are likely to improve DAP predictions significantly, which would keep their shown-to-becorrect distribution shape but would now match also in magnitude the dynamic response predictions.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

103

Table 9. Correction factor

Model 9-Storey Frame 20-Storey Frame

Interstorey Drift 0.75 0.62

Interstorey Shear 0.78 0.74

Interstorey Moment 0.99 0.74

As it might be observed in Fig. 65 to Fig. 68 (the full set of results is reported in Appendix G), such procedure leads to considerable improvements of results. Predictions provided by the modified adaptive scheme are very less biased and when the bias is large, as in the case of the interstorey overturning moment for the 20-Storey frame (Fig. 67), values are shifted towards the safety side with the tendency to overestimate results of dynamic analyses. Evidently, given that in this purely demonstrative example application, the results are being corrected with parameters that have been derived from the very same analyses, then the matching becomes quite impressive. Obviously, the proposed methodology will now need to be verified and calibrated by an extensive statistical-based parametric study that will consider a wide range of structural models and input motions.

9S LA

20S LA 20

18
8

TH
DAP
1st Mode
Uniform

16
7 6 Storey 5 4 3 2 1

14 12 Storey 10 8 6 4 2
0 2 4 6 Interstorey Drift [%] 8

Triangular
Code

2 4 Interstorey Drift [%]

Fig. 65. Corrected interstorey drift profile: 9-Storey frame (left) and 20-Storey frame (right)

Evaluation of Pushover Procedures for the Seismic Design of Buildings

104

9S LA 20 9 18 8 16 7 6 5 4 3 2 1 Storey Storey 14 12 10 8 6 4 2 0 0.5 1 1.5 Interstorey Drift 2 0 0.5

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

1 1.5 Interstorey Drift

Fig. 66. Corrected interstorey drift ratio (bias): 9-Storey frame (left) and 20-Storey frame (right)

9S LA 20 9 18 8 16 7 6 Storey 5 4 3 2 1 Storey 14 12 10 8 6 4 2 0 0.5 1 1.5 Interstorey Shear 2 0

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

0.5 1 1.5 Interstorey Moment

Fig. 67. Corrected bias in the interstorey shear for the 9-Storey frame (left) and bias in the interstorey moment for the 20-Storey frame (right)

Evaluation of Pushover Procedures for the Seismic Design of Buildings

105

9S LA 20 9 18 8 16 7 14 6 12 Storey 5 4 3 2 1 Storey 10 8 6 4 2 0 0.5 1 1.5 2 Interstorey Moment [kNm] 2.5 x 10

5

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

2000 4000 6000 8000 Interstorey Shear [kN]

Fig. 68. Corrected interstorey moment profile for the 9-Storey frame (left) and interstorey shear profile for the 20Storey frame (right)

5.3.5

Conclusions

Within this section the accuracy of the different static pushover methodologies in the prediction of the dynamic response of 2 high-rise prototype buildings has been evaluated for a set of 20 ground motions. Structural models, the 9- and 20-Storey LA frames developed for the SAC project, have been selected because they respond in the inelastic range for the records considered (as found in Goel and Chopra, 2004) and thus represent a severe test for the evaluation of the performance of static methodologies. Two separate studies have been performed: a preliminary one where the performance of DAP has been evaluated, and an extensive one where two modified version of the adaptive scheme, including an empirically-derived correction factors, have been proposed. Preliminary analyses shown that, even if estimates are not completely satisfactory, in particular for the 20-Storey frame, DAP, compared with conventional procedures, represents an alternative simpler procedure (involving a single pushover analysis) that allows to predict the response shape of tall steel buildings with very good accuracy, at least equal with respect to more complex procedures, such as MPA (Chopra and Goel, 2002). Since DAP predicts the shape of the drift profiles with superior accuracy in comparison to conventional pushover, but values are generally underestimated, two alternative methods have been

Evaluation of Pushover Procedures for the Seismic Design of Buildings

106

proposed. Such improved versions of DAP, without any modification of the adaptive algorithm, try to overcome the bias in DAP estimates. The first proposal consists in the adoption of a correction factor in the definition of the target displacement, whilst the second makes use of a scaling factor directly applied to DAP results. Both methods lead to promising results, improving estimates of deformations and internal forces with respect to preliminary analyses. However, such methods should represent only proposals with the aim to highlight how simple measures could lead to tremendous benefits in results, if parametric/statistical studies would lead to the conclusion that such a modified target displacement or such a scaling parameter could be derived.

6
6.1. Summary

Conclusions

The main objectives of the present work were to (I) evaluate and compare performances of traditional pushover schemes with those obtained with the more recently proposed Displacement-based Adaptive Pushover (DAP, Antoniou and Pinho, 2004(b)), (II) verify the applicability of an alternative single-step adaptive scheme which make use of an average spectrum, (III) further assess (with respect to Antoniou and Pinho (2004) and Lpez-Menjivar (2004)) the effectiveness of the three plot options (Maximum total drift vs. Correspondent base shear, Correspondent total drift vs. Maximum base shear, Maximum total drift vs. Maximum base shear) of IDA envelopes, and (IV) verify the reliability of DAP also for steel-frame structures. Nonlinear static procedures (NSPs) represents in the current design practice an easier and thus appealing alternative with respect to nonlinear dynamic analyses. They overcome the major drawbacks in time-history analyses, which in turn (a) require the simulation of an ensemble of site-specific ground motions compatible with the seismic hazard spectrum for the site, (b) they remain computationally demanding in particular when fibre-based structural analysis program are employed to model the seismic performance of large buildings, requiring 3D models with thousands of elements, considering also that a large number of analysis runs are required in order to get the average response of the structure. Traditionally, the increasing lateral load pattern used in pushover analysis has always been applied in invariant fashion, effectively implying that the response of a structure is controlled by a single fundamental mode shape that remains unchanged until collapse occurs. At most, two lateral load patterns, namely the first mode proportional and the uniform, were recommended to approximately bound the likely distribution of the inertia forces in the elastic and inelastic range, respectively. However, recent studies, some of which are summarised in the FEMA 440 report (ATC, 2005), raise

Evaluation of Pushover Procedures for the Seismic Design of Buildings

108

doubts on the effectiveness of conventional force-based static methods in estimating the seismic demand throughout the full deformation range: (i) inaccurate prediction of deformations when higher modes are important and/or the structure is highly pushed into its nonlinear post-yield range, (ii) inaccurate prediction of local damage concentrations, responsible for changing the modal response, (iii) inability of reproducing peculiar dynamic effects, neglecting sources of energy dissipation such as kinetic energy, viscous damping, and duration effects, (iv) difficulty in incorporating three-dimensional and cyclic earthquake loading effects. Thus, NSPs, due to their static nature, can only provide an estimate of the true dynamic response of the structure and thus the reliability of such design methods have to be carefully evaluated. Under this perspective it is essential that the pushover analysis, where an incremental-iterative solution of the static equilibrium equations is carried out to obtain the response of a structure subjected to a monotonically increasing lateral load pattern, which represents the fundamental basis of NSPs, could produce reliable estimates of the structural response at both global and local levels. In tandem with the present drive for performance-based seismic engineering, there is also currently a thrust for the development and code implementation of displacement or, more generally, deformationbased design and assessment methods. Therefore, it would seem that applying displacement loading in an adaptive fashion, rather than force actions, in pushover procedures would be an appropriate option for nonlinear static analysis of structures subjected to earthquake action. Hence, the performance of the Displacement-based Adaptive Pushover (DAP), recently proposed by Antoniou and Pinho (2004(b)), has been evaluated and compared with conventional force-based non-adaptive methodologies in the case of five steel frames. Two separate parametric studies have been carried out. Whilst the first follows the same outline of the research recently proposed in the FEMA 440 (ATC, 2005), the second parametric study considers two steel frames and the ensemble of ground motions developed for the FEMA-funded SAC joint venture project described in ATC-40 (Gupta and Krawinkler, 1999). In detail, in the first parametric study the attention has been focused on the performance of static procedures, in predicting both the global response, through a comparison of pushover curves with Incremental Dynamic Analysis (IDA) envelopes, as well as (ii) local response quantities, such as storey displacements, interstorey drifts, interstorey shears and overturning moments, in the case of higher mode effects at different levels of structural deformation. Instead, the second study concentrates on the evaluation of the performance of static procedures in predicting local response quantities in high-rise steel frames responding in the inelastic range.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

109

Under the perspective of the overall structural response, pushover curves provided by the different static methods are compared with Incremental Dynamic Analysis (IDA) envelopes. Results, also in the case of steel frames, confirm conclusions obtained in previous works by Antoniou and Pinho (2004(a)) and Papanikolaou et al. (2006), which indicated that for RC buildings the most effective way of plotting IDA envelopes should be Maximum Displacement vs. Corresponding Base Shear. In fact, the alternative options lead, in particular when higher modes responses are of relevance, to an unrealistic overestimation of the structural strength at large deformation levels. The capability of the static procedures to evaluate the global structural response in terms of its pushover curve can be considered satisfactory in the case of buildings where the first mode dominates the response, as in the case of the 3-Storey steel frame, in particular when the failure mechanism is well defined, such as for the weak-model. Instead, predictions became less accurate for the 9-Storey frame at large levels of structural deformations, due to the higher contribution of modal responses different from the first elastic mode, associated with the spread of inelastic deformation throughout the structure. In particular, the Uniform lateral load pattern provides a pushover curve closer to the IDA envelope in the case of the 3-Storey frame, whilst the other static schemes provide, with similar degree of precision, more accurate results in the case of the high-rise frame models. Among these, DAP shows higher consistency in leading to less erroneous predictions. However, any of the static procedures can capture the peculiar structural response for some NF records. Results provided by the static procedures at the local level (floor displacement, interstorey drift, interstorey shear and interstorey overturning moment), as come out from the first parametric study considering both the Ordinary Ground Motions (OGMs) and Near Field (NF) records responses of the building models, generally confirm what has been obtained in the ATC-55 project. In particular, the capacity of static procedures to predict the structural response in terms of floor displacements, and their limitation in the estimate of the other response quantities, in particular for high-rise buildings and large deformation levels. Results also confirm that the Uniform lateral load pattern leads in most of the cases to unrealistic results. Moreover, the current study shows that, even it cannot provide the optimal solution, the displacementbased adaptive scheme could represent an improvement with respect to the other static procedures. In fact, DAP provides, with a lower scatter along the building height, better estimates of interstorey drifts, shears and moments, in particular for high-rise buildings where the single mode schemes are fundamentally limited. It might also be noted that the dispersion of dynamic analyses results tends to increase with the magnitude of the deformation demand, in particular for what concern floor displacements and drifts,

Evaluation of Pushover Procedures for the Seismic Design of Buildings

110

and thus an adequate prediction of the response in these cases is more difficult, partially explaining the less accuracy of the estimates provided by the static methods. Results obtained in the second parametric study confirm the superiority of DAP against traditional non-adaptive methods also in predicting of the response of high-rise frames responding in the inelastic range, where higher-mode effects are of relevance. Hence, even if estimates are not always completely satisfactory, DAP, compared with conventional procedures, represents an alternative simpler procedure (involving a single pushover analysis) that allows to predict the response shape of tall steel buildings with very good accuracy, at least equal with respect to more complex procedures, such as MPA (Chopra and Goel, 2002). Moreover, the displacement-based adaptive methodology, not only preserve its reliability also in the case of structural models that experience roof drifts in the region of rapid decay of lateral capacity of the building (as for example the 20-Storey LA frame) where MPA fails, but also overcome possible drawbacks in the adoption of Modal Pushover Analyses. In fact, DAP is insensitive to peculiar structural responses that might induce reversal in the higher modes pushover curve, as for example occur in the MPA where particular countermeasures, such as the partial application of the total gravity loads (Goel and Chopra, 2004) or the assumption of a linear structural response in the higher modes (Goel and Chopra, 2006), have to be accounted for. The preliminary stage of analysis in this second parametric study clearly shown that DAP improves results against non-adaptive procedures, in particular in the prediction of the shape of the drift profiles of high-rise steel frames. This characteristic of DAP is particularly positive and encouraging, since interstorey drift profiles provide valuable information on the yield/failure mechanism that the structure will develop and can also be directly related to structural and non-structural damage. However, the actual values of drifts are slightly underestimated by the adaptive pushover scheme. Hence, whilst the under-prediction of the actual values can be overcome with relative ease by the introduction of empirically-derived factors, the correct prediction of the interstorey drift profile requires an algorithm that is able to take due account of both higher mode contributions and distribution of inelastic deformations among structural members. Thus, two alternative modified DAP procedures, based on empirical observations, have been proposed in order to overcome such underperformance. The first, account for the bias in the interstorey drift estimates by mean of a new target displacement in the analysis (named, Displacement correction factor method). In the second, instead, results obtained in DAP analyses are directly corrected through the adoption of a proper scaling factor (named, Scaling factor method).

Evaluation of Pushover Procedures for the Seismic Design of Buildings

111

Both methods lead to promising results, improving estimates of deformations and internal forces with respect to preliminary analyses. However, such methods should represent only proposals with the aim to highlight how simple measures could lead to tremendous benefits in results, if parametric/statistical studies would lead to the conclusion that such a modified target displacement or such a scaling parameter could be derived. Furthermore, the structural analysis fiber-based program SeismoStruct has been verified for steel frame systems. Results, obtained comparing analytical estimates with results on a full scale structure (Matsumiya et al., 2004 and Nakashima et al., 2006), confirm the reliability of the software also in the case of steel moment-resisting frames. Finally, it has been shown that the alternative adaptive scheme DAP-AS tested within the first parametric study, which make use of an average spectrum (design spectrum) for the scaling of the displacement load pattern, allows for estimates of local response quantities close to the mean DAP estimates. This allows for an easier single-step application of such pushover scheme accounting for the specific design spectrum properties. In fact, since DAP-AS provides essentially results coincident with the average among the record-specific DAP estimates, the design procedure, adopting an average spectrum scaling, can be performed within a single-run analysis. Thus, it might be concluded that results obtained in this study seem to indicate that, when compared with other pushover procedures, DAP leads to higher accuracy in the prediction of global and local response parameters of steel buildings, particularly in those cases where the influence of higher modes of vibration is important. It is also shown that the employment of an average response spectral shape leads to satisfactory results, thus rendering the procedure very much applicable within a design application framework, where standard code spectral shapes are prescribed. In addition, it is also noted that the additional modelling and computational effort requested to run such type of analysis is, with respect to conventional pushover procedures, negligible; a single-pushover needs to be run and no special results manipulation is required.

6.2.

Future Research

The Displacement-based Adaptive Pushover has been implemented adopting the Complete Quadratic Combination rule (CQC). As described by Kunnath (2004) and Lpez-Menjivar (2004), the adoption of a quadratic combination rule (CQC or SRSS) in computing the adaptive updating of the loading vector still represent the main reason for the underperformance of multimodal procedures. In fact, these rules

Evaluation of Pushover Procedures for the Seismic Design of Buildings

112

will inevitably lead to monotonically increasing load vectors, because the sign change in applied loads at any location is precluded, whilst it may be needed to represent the uneven redistribution of forces after an inelastic mechanism is triggered at some location. Few researchers (among these, Matsumori et al., 1999; Kunnath, 2004; Lpez-Menjivar, 2004) proposed alternative combination rules, and additional studies are needed about this topic. Moreover, more detail parametric/statistical studies should investigate about the possibility that such a modified target displacement or such a scaling parameter, here only suggested as possible improvements, could be derived Furthermore, since all the tests has been carried out on 2D models, it is needed that the static as well adaptive procedures are tested on 3D space structures preferably on irregular plan buildings.

References
Albanesi T., Biondi S. and Petrangeli M., 2002, Pushover analysis: An energy based approach, Proceedings of the 12th European Conference on Earthquake Engineering, Paper 605. Elsevier Science Ltd. Antoniou S. and Pinho R., 2004(a), Advantages and limitations of adaptive and non-adaptive force-based pushover
procedures, Journal of Earthquake Engineering, Vol. 8, No. 4, pp. 497-522.

Antoniou S. and Pinho R., 2004(b), Development and verification of a displacement-based adaptive pushover procedure, Journal of Earthquake Engineering, Vol. 8, No. 5, pp. 643-661. ASCE, 2000, Prestandard and commentary for the seismic rehabilitation of buildings, FEMA 356 Report, prepared by the American Society of Civil Engineers for the Federal Emergency Management Agency, Washington, D.C. ATC, 1996, Seismic evaluation and retrofit of concrete buildings, ATC-40 Report, Volumes 1and 2, Applied Technology Council, Redwood City, California. ATC, 2005, Improvement of nonlinear static seismic analysis procedures, FEMA 440 Report, Applied Technology Council, Redwood City, California. Aydinoglu M.N., 2003, An incremental response spectrum analysis procedure based on inelastic spectral deformation for
multi-mode seismic performance evaluation, Bulletin of Earthquake Engineering, No. 1, pp. 3-36.

Ballio G. and Mazzolani F.M., 1987, Strutture in acciaio, Ulrico Hoepli editore S.p.A., Milano. Bommer J.J. and Martinez-Pereira A., 1999, The effective duration of earthquake strong motion, Journal of Earthquake Engineering, Vol. 3, No. 2, pp. 127-172. Bommer J.J. and Acevedo A.B., 2004, The use of real earthquake accelerograms as input to dynamic analysis, Journal of Earthquake Engineering, Vol. 8 (Special Issue 1), pp. 43-91. Bracci J.M., Kunnath S. K. and Reinhorn A.M., 1997, Seismic performance and retrofit evaluation of reinforced
concrete structures, Journal of Structural Engineering, Vol. 123, No. 1, pp. 3-10.

Casarotti C., 2005, Adaptive pushover-based methods for seismic assessment and design of bridge structures, ROSE School, PhD Dissertation. Chintanapakdee C. and Chopra A.K., 2003, Evaluation of modal pushover analysis using generic frames, Earthquake Engineering and Structural Dynamics, Vol. 32, Wiley Publishers, Hoboken, New Jersey, pp.417-442.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

114

Chopra A.K. and Goel R.K., 1999, Capacity-demand-diagram methods based on inelastic design spectrum, Earthquake Spectra, Vol. 15, No. 4, pp. 637-656. Chopra A.K. and Goel R.K., 2001, A modal pushover analysis procedure to estimate seismic demands for buildings:
Theory and preliminary evaluation, PEER Report 2001/03, Pacific Earthquake Engineering Research Center, College

of Engineering, University of California, Berkeley, January. Chopra A.K. and Goel R.K., 2002, A modal pushover analysis procedure for estimating seismic demands for buildings, Earthquake Engineering and Structural Dynamics, Vol. 31, pp.561-582. Chopra, A.K. and Goel, R.K. 2006, Evaluation of the Modal Pushover Analysis Procedure for Unsymmetric-Plan
Buildings, Proceedings of the 8th U.S. Conference on Earthquake Engineering, San Francisco, California, 17-21

April 2006. Der Kiureghian A., 1981, A response spectrum method for random vibration analysis of MDOF systems, Earthquake Engineering and Structural Dynamics, No 9, pp.419-435. Elnashai A.S., 2001, Advanced inelastic static (pushover) analysis for earthquake applications, Structural Engineering and Mechanics, Col. 12, No. 1, pp. 51-69. Fajfar P., 1999, Capacity spectrum method based on inelastic demand spectra, Earthquake Engineering and Structural Dynamics, No. 28, pp. 979-993. Fishinger M., Beg D., Isakovi T.A., Tomaevi M. and arni R., 2004, Performance-based assessment from
general methodologies to specific implementations, International Workshop on Performance-based Seismic Design, Bled

June 28/July 1. Freeman S.A., 1994, The capacity spectrum method for determining the demand displacement, ACI Spring Convention. Goel R.K. and Chopra A.K., 2004, Evaluation of modal and FEMA pushover Analysis: SAC Buildings, Earthquake Spectra, Vol. 20, No. 1, Earthquake Engineering Research Institute, Oakland, California, pp.225-254. Goel R.K. and Chopra A.K., 2005(a), Response to Maisons discussion of Evaluation of Modal ans FEMA
Pushover Analysis: SAC Buildings, Earthquake Spectra, Vol. 21, No. 1, pp. 277-279.

Goel R.K. and Chopra A.K., 2005(b), Role of higher-mode pushover analyses in seismic analysis of buildings, Earthquake Spectra, Vol. 21, No. 4, pp. 1027-1041. Guedes J., 1997, Seismic behaviour of reinforced concrete bridges. Modeling, numerical analysis and experimental
assessment, Ph.D. Thesis, Univ. of Porto, Porto, Portugal.

Gulkan P. and Sozen M.A., 1974, Inelastic response of reinforced concrete structures to earthquake motions, ACI Journal, Vol. 71, pp. 604-610. Gupta A. and Krawinkler H., 1999, Seismic demands for performance evaluation of steel moment resisting frame
structures (SAC Task 5.4.3), Report No. 132, John A. Blume Earthquake Engineering Center, Stanford University,

Stanford, California. Gupta B. and Kunnath S.K., 2000, Adaptive spectra-based pushover procedure for seismic evaluation of structures, Earthquake Spectra, Vol. 16, No. 2, pp. 367-391. Hall J.F., 2006, Problems encountered from the use (or misuse) of Rayleigh damping, Earthquake Engineering and Structural Dynamics, Vol. 35, Issue 5, pp. 525-545

Evaluation of Pushover Procedures for the Seismic Design of Buildings

115

Hamburger R.O., Foutch D.A. and Cornell C.A., 2000, Performance basis of guidelines for evaluation, upgrade and
design of moment-resisting steel frames, Proceedings of the Twelfth World Conference on Earthquake Engineering,

Auckland, New Zealand, paper No. 2543. Hernndez-Montes E., Kwon O-S and Ascheim M., 2004, An energy-based formulation for first- and multiplemode nonlinear static (pushover) analyses, Journal of Earthquake Engineering, Vol. 8, No.1, pp.69-88.

Hughes T.J.R., 1987, The finite element method, linear static and dynamic finite element analysis, Prentice-Hall, Inc. Iwan W.D., 1980, Estimating inelastic response spectra from elastic response spectra, Earthquake Engineering and Structural Dynamics, No. 8, pp. 375-388. Iwan W.D., 1999, Implications of near-fault ground motions for structural design, US-Japan workshop on Performance-Based Earthquake Engineering methodology for RC building structures, Maui, Hawaii, (from PEER, UC, Berkeley). Izzuddin B.A., 2001, Conceptual issues in geometrically nonlinear analysis of 3-D frame structures, Computer Methods in Applied Mechanics and Engineering, Vol. 191, pp. 1029-1053. Kim S. and DAmore E., 1999, Push-over analysis procedure in earthquake engineering, Earthquake Spectra, Vol. 15, Issue 3, pp. 417-434. Krawinkler H. and Seneviratna G.D.P.K., 1998, Pros and cons of a pushover analysis of seismic performance
evaluation, Engineering Structures, Vol. 20, No. (4-6), pp. 452-464.

Kunnath S.K., 2004, Identification of modal combination for nonlinear static analysis of building structures, ComputerAided Civil and Infrastructure Engineering, Vol. 19, pp 246-259. Lawson R.S., Vance V. and Krawinkler H., 1994, Nonlinear static push-over analysis - why, when, and how?, Proceedings of the Fifth U.S. National Conference on Earthquake Engineering, Earthquake Engineering Research Institute, Oakland, California, I, Vol. I, pp. 283-292. Lefort T., 2000, Advance pushover analysis of RC multi-storey buildings, Engineering Seismology and Earthquake Engineering Section, Imperial College of Science, Technology and Medicine. Report No. 91/2, March 1991. Lpez-Menjivar M.A., 2004, Adaptive pushover of 2-D reinforced concrete buildings, Ph.D. Thesis, European School for Advanced Studies in Reduction of Seismic Risk (ROSE School), University of Pavia, Italy. Maison B., 2005, Discussion of evaluation of modal and FEMA pushover analysis: SAC buildings, Earthquake Spectra, Vol. 21, No.1, pp. 275-275. Matsumiya T., Suita K., Chusilp P., Nakashima M., 2004, Full-scale test of three-story steel moment frames for
examination of extremely large deformation and collapse behaviour, Proceedings of the 13th World Conference on

Earthquake Engineering, Vancouver, Paper No. 3471, 16 August 2004. Matsumori T., Otani S., Shioara H. and Kabeyasawa T., 1999, Earthquake member deformation demands in
reinforced concrete frame structures, Proceedings of the U.S.-Japan Workshop on Performance-Based Earthquake

Engineering Methodology for R/C Building Structures, PEER Center Report, UC Berkeley, pp.79-94, Maui, Hawaii. Meireles H., Pinho R., Bento R. and Antoniou S., 2006, Verification of an adaptive pushover technique for the 3D
case, Proceedings of the 13th European Conference on Earthquake Engineering, Geneva, Switzerland, Paper 619.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

116

Miranda E., 2000, Inelastic displacement ratios for structures on firm sites, Journal of Structural Engineering, No. 126, pp. 1150-1159. Mogadam A.S. and Tso W.K., 2002, A pushover procedure for tall buildings, Proceedings of the 12th European Conference on Earthquake Engineering, Paper 395, Elsevier Science Ltd. Mwafy A.M. and Elnashai S.A., 2000, Static pushover versus dynamic-to-collapse analysis of RC buildings, Engineering Seismology and Earthquake Engineering Section, Imperial College of Science, Technology and Medicine. Report No. 00/1, January. Naeim F. and Lobo R, 1999, Avoiding common pitfalls in pushover analysis, Proceedings of 8th Canadian Conference on Earthquake Engineering, Vancouver, pp. 269-274. Nakashima M., Matsumiya T., Suita K., Liu D., 2006, Test on full-scale three-story steel moment frame and
assessment of ability of numerical simulation to trace cyclic inelastic behaviour, Earthquake Engineering and Structural

Dynamics, No. 35, pp. 3-19. Newmark N.M. and Hall W.J., 1982, Earthquake spectra and design, Earthquake Engineering Research Institute, Berkeley, CA. Newmark N.M., 1959, A method for computation for structural dynamics, ASCE, Journal of the Engineering and Mechanics Division, Vol. 85, Issue EM3, pp.67-94. Ohtori Y., Christenson R.E., Spencer B.F. and Dyke S.J., 2003, Benchmark problems in seismically excited
nonlinear buildings, Journal of Engineering Mechanics, ASCE, 129, to appear, http://cee.uiuc.edu/sstl/.

Papanikolaou V.K., Elnashai A.S. and Pareja J.F., 2006, Evaluation of conventional and adaptive pushover analysis, Journal of Earthquake Engineering, Vol. 10, No. 1, pp. 127-132. Paret T.F., Sasaki K.K., Elibeck D.K. and Freeman S.A., 1996, Approximate inelastic procedures to identify
failure mechanism from higher mode effects, Proceedings of the 11th World Conference on Earthquake Engineering,

Paper 966, Pergamon, Elsevier Science Ltd, Acapulco, Mexico. Pinto A.V., Verzeletti G., Pegon P., Magonette G., Negro P. and Guedes J., 1996, Pseudo-dynamic testing of
large-scale R/C bridges, ELSA Lab., Report EUR 16378 EN.

prENV 1998-1, 1994, Eurocode 8: design provisions for earthquake resistance of structures Part 1: General rules. Priestley M.J.N. and Grant D.N., 2005, Viscous damping in seismic design and analysis, Journal of Earthquake Engineering, Vol.9 (Special Issue 2), pp. 229-255. Priestley M.J.N. and Kowalsky M.J., 2000, Direct displacement-based design of concrete buildings, Bulletin of Earthquake Engineering, Vol.33, No. 4. Priestley M.J.N., 1993, Myths and fallacies in earthquake engineering-conflicts between design and reality, Bulletin of New Zealand National Society for Earthquake Engineering, Vol. 26, No. 3, pp. 329-341. Reinhorn A., 1997, Inelastic analysis techniques in seismic evaluations, in seismic design methodologies for the next
generation of codes, Krawinkler and Fajfar (editors), Balkema, 277-287.

Requena M. and Ayala G., 2000, Evaluation of a simplified method for the determination of the non-linear seismic
response of RC frames, Proceedings of the Twelfth World Conference on Earthquake Engineering, Paper 2109.

New Zealand Society for Earthquake Engineering, Upper Hutt, New Zealand.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

117

Rosenblueth E. and Herrera I., 1964, On a kind of hysteretic damping, Journal of Engineering Mechanics Division, ASCE, Vol.90, pp.37-48. Sasaki K.K., Freeman S.A. and Paret T.F., 1998, Multimode pushover procedure (MMP) a method to identify the
effects of higher modes in a pushover analysis, Proceedings of the 6th U.S. National Conference on Earthquake

Engineering, Earthquake Engineering Research Institute, Oakland, California. Seismosoft, 2005, SeismoStruct A computer program for static and dynamic nonlinear analysis of framed structures, [online], available from URL: http//www.seismosoft.com. Seneviratna G.D.P.K., 1995, Evaluation of inelastic MDOF effects for seismic design, Ph.D. Dissertation, Dept. of civil Engineering, Standford University. Sommerville P., Smith N., Punyamurthula S. and Sun J., 1997, Development of ground motion time histories for
phase 2 of the FEMA/SAC steel project, SAC Background Document, Report No. SAC/BD/-7/04, Applied

Technology Council, Redwood City, CA. Valles R.E., Reinhorn A.M., Kunnath S.K., Li C. and Madan A., 1996, IDARC2D Version 4.0: A computer
program for the inelastic analysis of buildings, Technical Report NCEER 96-0010, State University of New York at

Buffalo, Buffalo, NY. Vamvatsikos D. and Cornell C.A., 2002, Incremental dynamic analysis, Earthquake Engineering Structural Dynamics, Vol. 31, No. 3, pp. 491-514. Yu K., Heintz J. and Poland C., 2001, Assessment of nonlinear static procedures for seismic evaluation of building
structures, Proceedings US-Japan Joint Workshop and Third Grantees Meeting, U.S.-Japan cooperative research

on urban earthquake disaster mitigation, 15-16 August 2001, University of Washington, Seattle, Washington, pp.431-450.

Evaluation of Pushover Procedures for the Seismic Design of Buildings

118

Appendix A: Prototype Buildings (members section and frame dynamic properties)

A.1 Section properties Table 10. Steel section properties Section W36x160 W36x135 W33x118 W30x116 W30x108 W30x99 W27x84 W24x335 W24x229 W24x192 W24x131 W24x117 W24x84 W24x68 W24x62 W21x50 W18x35 W16x26 Area [m2] 0.030323 0.025613 0.022387 0.022064 0.02 0.018774 0.016000 0.06 0.04 0.04 0.02 0.02 0.02 0.012968 0.01 0.01 0.006645 0.004955 d [m] 0.915 0.903 0.835 0.762 0.76 0.753 0.678 0.70 0.66 0.65 0.62 0.62 0.61 0.603 0.60 0.53 0.450 0.399 tw [m] 0.017 0.015 0.014 0.014 0.01 0.013 0.012 0.04 0.02 0.02 0.02 0.01 0.01 0.011 0.01 0.01 0.008 0.006 bf [m] 0.305 0.304 0.292 0.267 0.27 0.265 0.253 0.34 0.33 0.33 0.33 0.33 0.23 0.228 0.18 0.17 0.152 0.140 tf [m] 0.026 0.020 0.019 0.022 0.02 0.017 0.016 0.06 0.04 0.04 0.02 0.02 0.02 0.015 0.01 0.01 0.011 0.009

Evaluation of Pushover Procedures for the Seismic Design of Buildings

120

W14x500 W14x455 W14x370 W14x311 W14x283 W14x257 W14x233 W14x68 W14x22

0.094839 0.086451 0.070322 0.058968 0.053742 0.048774 0.044193 0.012903 0.00

0.498 0.483 0.455 0.435 0.425 0.416 0.407 0.357 0.35

0.056 0.051 0.042 0.036 0.033 0.030 0.027 0.011 0.01

0.432 0.428 0.418 0.412 0.409 0.406 0.404 0.255 0.13

0.089 0.082 0.068 0.057 0.053 0.048 0.044 0.018 0.01

Square Hallow width [m] Section 15x15x2 15x15x1.25 15x15x1 15x15x0.75 15x15x0.5 0.381 0.381 0.381 0.381 0.381

depth [m] 0.381 0.381 0.381 0.381 0.381

thickness [m] 0.051 0.032 0.025 0.019 0.0127

Fig. 69. Steel cross section parameters

Evaluation of Pushover Procedures for the Seismic Design of Buildings

121

A.2

Dynamic properties Table 11. Building models: first three periods of vibration

Building Model 3S 9S 20S

T1 [s] 1.0187 2.248 3.79

T2 [s] 0.330 0.847 1.332 Table 12. 3-Storey frame: mode shapes.

T3 [s] 0.173 0.484 0.772

Level 1 2 3

1st mode 0.00 0.27 0.66

2nd mode -1.19 -1.13 1.00 Table 13. 9-Storey frame: mode shapes.

3rd mode 3.95 -3.25 1.00

Level 1 2 3 4 5 6 7 8 9

1st mode 0.17 0.28 0.39 0.50 0.61 0.71 0.81 0.92 1.00

2nd mode -0.41 -0.62 -0.76 -0.80 -0.70 -0.47 -0.06 0.49 1.00

3rd mode 0.92 1.19 0.97 0.33 -0.51 -1.16 -1.14 -0.27 1.00

Evaluation of Pushover Procedures for the Seismic Design of Buildings

122

Table 14. 20-Storey frame: mode shapes. Level 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1st mode 0.07 0.13 0.18 0.24 0.30 0.36 0.41 0.47 0.52 0.58 0.63 0.68 0.73 0.78 0.82 0.87 0.90 0.94 0.97 1.00 2nd mode -0.20 -0.34 -0.47 -0.59 -0.68 -0.74 -0.78 -0.78 -0.74 -0.68 -0.58 -0.45 -0.30 -0.13 0.07 0.26 0.46 0.66 0.84 1.00 3rd mode 0.30 0.50 0.65 0.73 0.73 0.65 0.49 0.27 0.03 -0.22 -0.46 -0.64 -0.75 -0.76 -0.66 -0.45 -0.15 0.22 0.63 1.00

Appendix B: First parametric study scaling factors

Table 15. 3-Storey frame: scaling factors for OGM Drift [%] 0.5 2 4
0.3035 1.5102 3.2663 0.3552 1.6942 2.4675 0.295 1.1609 2.2117 0.4386 1.3813 1.9482 0.2964 1.3617 2.2937 0.2206 1.3828 1.9177 0.2157 1.7014 2.8122 0.5719 2.4509 3.189 0.4303 2.6689 4.4371 0.4377 2.724 5.7964 0.5557 2.0373 3.0931

A1

A2

A3

A4

A5

A6

A7

A8

A9

A10

A11

Table 16. 3-Storey weak-frame: scaling factors for OGM Drift [%] 0.5 2 4
0.361 1.469 3.5973 0.4029 1.4752 2.2217 0.3129 1.3836 2.0269 0.4238 1.2567 1.9283 0.3117 1.2142 2.4099 0.2091 1.1927 2.3512 0.1943 1.3522 2.1092 0.5731 2.1715 2.9986 0.4073 2.4012 4.4596 0.4134 2.8274 4.5558 0.5517 1.853 3.9173

A1

A2

A3

A4

A5

A6

A7

A8

A9

A10

A11

Table 17. 9-Storey frame: scaling factors for OGM Drift [%] 0.5 2 4
0.4235 2.4419 4.6059 0.277 2.234 3.954 0.5597 3.5571 5.4052 0.3303 2.6581 3.2483 0.3091 3.1579 5.1544 0.1508 1.4181 3.4181 0.1812 1.3428 3.1662 0.2543 2.7622 8.3254 0.3155 3.4853 8.8455 0.3544 4.1952 6.5388 0.8223 2.7019 6.8514

A1

A2

A3

A4

A5

A6

A7

A8

A9

A10

A11

Evaluation of Pushover Procedures for the Seismic Design of Buildings

124

Table 18. 9-Storey weak-frame: scaling factors for OGM Drift [%] 0.5 2 2.7
0.3955 2.8924 3.6901 0.255 2.3921 3.230 0.5407 3.6139 4.721 0.2453 3.1944 3.418 0.3736 3.1623 3.877 0.1608 1.4242 1.9801 0.1812 1.6088 1.9302 0.2543 2.1757 3.408 0.3155 2.9551 3.4551 0.3544 4.2568 5.864 0.8223 4.6055 5.5054

A1

A2

A3

A4

A5

A6

A7

A8

A9

A10

A11

Appendix C: First parametric study - Capacity Curves

C.1 OGMs

C.1.1 Representations of IDA envelopes

7000 6000 5000 4000 3000 2000 1000 0 0.00

Mean DAP mean+st.dv mean-st.dv 1st Mode Uniform Triangular Code Max. Drift vs. Corr. Base Shear Max. Drift vs. Max. Base Shear Corr. Drift vs. Max. Base Shear

Base Shear [kN]

0.10

0.20

0.30

0.40

0.50

Top Floor Displacement [m]

Fig. 70. 3-Storey frame

Evaluation of Pushover Procedures for the Seismic Design of Buildings

126

4000 3500 3000

Mean DAP mean+st.dv mean-st.dv 1st Mode Uniform Triangular Code Max. Drift vs. Corr. Base Shear Max. Drift vs. Max. Base Shear Corr. Drift vs. Max. Base Shear

Base Shear [kN]

2500 2000 1500 1000 500 0 0.00

0.10

0.20

0.30

0.40

0.50

Top Floor Displacement [m]

Fig. 71. 3-Storey weak frame

12000 10000

Mean DAP mean+st.dv mean-st.dv 1st Mode Uniform

Base Shear [kN]

8000 6000 4000 2000 0 0.00

Triangular Code Max. Drift vs. Corr. Base Shear Max. Drift vs. Max. Base Shear Corr. Drift vs. Max. Base Shear

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

Top Floor Displacement [m]

Fig. 72. 9-Storey frame

9000 8000 7000
Mean DAP mean+st.dv mean-st.dv 1st Mode Uniform Triangular Code Max. Drift vs. Corr. Base Shear Max. Drift vs. Max. Base Shear Corr. Drift vs. Max. Base Shear

Base Shear [kN]

6000 5000 4000 3000 2000 1000 0 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60

Top Floor Displacement [m]

Fig. 73. 9-Storey weak frame

Evaluation of Pushover Procedures for the Seismic Design of Buildings

127

C.1.2 Pushover curves

6000 5000
Mean DAP mean+st.dv mean-st.dv

Base Shear [kN]

4000 3000 2000 1000 0 0.00

1st Mode Uniform Triangular Code DAP-AS Max. Drift vs. Corr. Base Shear Mean+St.dev. Mean-St.dev.

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

Top Floor Displacement [m]

Fig. 74. 3-Storey frame

4000 3500 3000
Mean DAP mean+st.dv mean-st.dv 1st Mode Uniform Triangular Code DAP-AS Max. Drift vs. Corr. Base Shear Mean+St.dev. Mean-St.dev.

Base Shear [kN]

2500 2000 1500 1000 500 0 0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

Top Floor Displacement [m]

Fig. 75. 3-Storey weak frame

9000 8000 7000
Mean DAP mean+st.dv mean-st.dv 1st Mode Uniform Triangular Code Max. Drift vs. Corr. Base Shear Mean+St.dev. Mean-St.dev.

Base Shear [kN]

6000 5000 4000 3000 2000 1000 0 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60

Top Floor Displacement [m]

Fig. 76. 9-Storey frame

Evaluation of Pushover Procedures for the Seismic Design of Buildings

128

8000 7000
Mean DAP

6000

mean+st.dv mean-st.dv 1st Mode Uniform Triangular Code DAP-AS Max. Drift vs. Corr. Base Shear Mean+St.dev. Mean-St.dev.

Base Shear [kN]

5000 4000 3000 2000 1000 0 0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

Top Floor Displacement [m]

Fig. 77. 9-Storey weak frame

C.1.3 Mean and maximum values of E1

Errors-E1
100.00% 90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% DAP 1st Mode Uniform Triangular Code 100.00% 90.00% 80.00% 70.00% 60.00%

Errors-E2

Mean Max

50.00% 40.00% 30.00% 20.00% 10.00% 0.00% DAP 1st Mode Uniform Triangular Code

Mean Max

E1

Pushover

E1

Pushover

3-Storey frame
Errors-E1
100.00% 90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% DAP 1st Mode Uniform Triangular Code 100.00% 90.00% 80.00% 70.00% 60.00%

3-Storey weak frame

Errors-E1

Mean Max

50.00% 40.00% 30.00% 20.00% 10.00% 0.00% DAP 1st Mode Uniform Triangular Code

Mean Max

E1

Pushover

E1

Pushover

9-Storey frame

9-Storey weak frame Fig. 78. E1-Mean and Maximum values

Evaluation of Pushover Procedures for the Seismic Design of Buildings

129

C.1.4 E2
Errors-E2
1.60 1.40 1.20 1.00 1.60 1.40 1.20

Errors-E2

0.80 0.60 0.40 0.20 0.00 0.000

DAP 1st Mode Uniform Triangular Code

1.00

0.80 0.60 0.40 0.20 0.00 0.000

DAP 1st Mode Uniform Triangular Code

E2

0.050

0.100

0.150

0.200

0.250

0.300

0.350

0.400

0.450

0.500

E2

0.100

0.200

0.300

0.400

0.500

0.600

Top Floor Displacement TH [m]

Top Floor Displacement TH [m]

3-Storey frame
Errors-E2
1.60 1.40 1.20 1.00 1.60 1.40 1.20

3-Storey weak frame

Errors-E2

0.80 0.60 0.40 0.20 0.00 0.000

DAP 1st Mode Uniform Triangular Code

1.00

0.80 0.60 0.40 0.20 0.00 0.000

DAP 1st Mode Uniform Triangular Code

E2

0.200

0.400

0.600

0.800

1.000

1.200

1.400

1.600

E2

0.200

0.400

0.600

0.800

1.000

1.200

Top Floor Displacement TH [m]

Top Floor Displacement TH [m]

9-Storey frame Fig. 79. E2

9-Storey weak frame

Evaluation of Pushover Procedures for the Seismic Design of Buildings

130

C.2

NF records

C.2.1 3-Storey frame

8000 7000 6000 5000 4000 3000 2000 1000 0 0 0.1 0.2 0.3 Top Floor Displacement [m] 0.4 0.5 0.6 Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

Base Shear [kN]

Fig. 80. 3-Storey frame-NF records-Capacity Curves ERZMV1

7000

6000

5000 Base Shear [kN]

4000

3000

2000

Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

1000

0 0 0.1 0.2 0.3 Top Floor Displacement [m] 0.4 0.5 0.6

Fig. 81. 3-Storey frame-NF records-Capacity Curves LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

131

7000

6000

5000 Base Shear [kN]

4000

3000

2000

Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

1000

0 0 0.1 0.2 0.3 Top Floor Displacement [m] 0.4 0.5 0.6

Fig. 82. 3-Storey frame-NF records-Capacity Curves RRSMV1

8000 7000 6000 5000 4000 3000 2000 1000 0 0 0.1 0.2 0.3 Top Floor Displacement [m] 0.4 0.5 0.6 Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

Base Shear [kN]

Fig. 83. 3-Storey frame-NF records-Capacity Curves SCHMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

132

E1-ERZMV1
40.00% 35.00% 30.00% 25.00%
Mean Max

E1-LUCMV1

40.00% 35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00% 0.00% Uniform Triangular 1st Mode Code DAP

20.00% 15.00% 10.00% 5.00% 0.00% Uniform Triangular 1st Mode Code DAP

Mean Max

E1-RRSMV1
40.00% 35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00% 0.00% Uniform Triangular 1st Mode Code DAP Mean Max
40.00% 35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00% 0.00% Uniform Triangular

E1-SCHMV1

Mean Max

1st Mode

Code

DAP

Fig. 84. 3-Storey frame-NF records-E1

E2-ERZMV1
1.20 1.60 1.40 1.20 0.80 Triangular 1st Mode Code DAP Uniform 1.00 0.80 0.60 0.40 0.20 0.20 0.00 0 0.1 0.2 0.3 Top Floor TH displacement [m] 0.4 0.5 0.6 0.00 0 0.05 0.1 0.15 0.2

E2-LUCMV1

1.00

0.60

0.40

Triangular 1st Mode Code DAP Uniform

0.25

0.3

0.35

0.4

0.45

Top Floor TH displacement [m]

E2-RRSMV1
1.40 1.20 1.00 0.80 0.60 0.40 0.20 0.00 0 0.1 0.2 0.3 Top Floor TH displacement [m] 0.4 0.5 0.6 Triangular 1st Mode Code DAP Uniform 1.00 0.80 0.60 0.40 0.20 0.00 0 0.1 0.2 1.60 1.40 1.20

E2-SCHMV1

Triangular 1st Mode Code DAP Uniform

0.3 Top Floor TH displacement [m]

0.4

0.5

0.6

Fig. 85. 3-Storey frame-NF records-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

133

C.2.2 3-Storey weak frame

4000 3500 3000 2500 2000 1500 1000 500 0 0 0.1 0.2 0.3 Top Floor Displacement [m] 0.4 0.5 0.6 Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

Base Shear [kN]

Fig. 86. 3-Storey weak frame-NF records-Capacity Curves ERZMV1

4000 3500 3000 2500 2000 1500 1000 500 0 0 0.1 0.2 0.3 Top Floor Displacement [m] 0.4 0.5 0.6 Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

Base Shear [kN]

Fig. 87. 3-Storey weak frame-NF records-Capacity Curves LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

134

4000 3500 3000 2500 2000 1500 1000 500 0 0 0.1 0.2 0.3 Top Floor Displacement [m] 0.4 0.5 0.6 Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

Base Shear [kN]

Fig. 88. 3-Storey weak frame-NF records-Capacity Curves RRSMV1

4000 3500 3000 2500 2000 1500 1000 500 0 0 0.1 0.2 0.3 Top Floor Displacement [m] 0.4 0.5 0.6 Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

Base Shear [kN]

Fig. 89. 3-Storey weak frame-NF records-Capacity Curves SCHMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

135

E1-ERZMV1

E1-LUCMV1

40.00% 35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00% 0.00% Uniform Triangular 1st Mode Code DAP Mean Max

40.00% 35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00% 0.00% Uniform Triangular 1st Mode Code DAP Mean Max

E1-RRSMV1
40.00% 35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00% 0.00% Uniform Triangular 1st Mode Code DAP Mean Max

E1-SCHMV1

40.00% 35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00% 0.00% Uniform Triangular 1st Mode Code DAP Mean Max

Fig. 90. 3-Storey-weak frame-NF records-E1

E2-ERZMV1
1.20 1.60 1.40 1.20 0.80 Triangular 1st Mode Code DAP Uniform 1.00 0.80 0.60 0.40 0.20 0.20 0.00 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 Top Floor TH displacement [m] 0.00 0 0.05 0.1 0.15 0.2

E2-LUCMV1

1.00

0.60

0.40

Triangular 1st Mode Code DAP Uniform

0.25

0.3

0.35

0.4

0.45

0.5

Top Floor TH displacement [m]

E2-RRSMV1
1.14 1.12 1.10 1.08 1.06 1.04 1.02 1.00 0.98 0.96 0.94 0 0.1 0.2 0.3 Top Floor TH displacement [m] 0.4 0.5 0.6
0.00 0 0.05 0.1 0.15 0.2 1.00 1.20

E2-SCHMV1

Triangular 1st Mode Code DAP Uniform

0.80

0.60

0.40

Triangular 1st Mode Code DAP Uniform

0.20

0.25

0.3

0.35

0.4

0.45

0.5

Top Floor TH displacement [m]

Fig. 91. 3-Storey-weak frame-NF records-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

136

C.2.3 9-Storey frame

12000

10000 Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

8000 Base Shear [kN]

6000

4000

2000

0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Top Floor Displacement [m]

Fig. 92. 9-Storey frame-NF records-Capacity Curves ERZMV1

12000

10000

8000 Base Shear [kN]

6000

4000

Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

2000

0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Top Floor Displacement [m]

Fig. 93. 9-Storey frame-NF records-Capacity Curves LUCMV11

Evaluation of Pushover Procedures for the Seismic Design of Buildings

137

14000

12000

10000 Base Shear [kN]

8000

6000

4000

Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

2000

0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Top Floor Displacement [m]

Fig. 94. 9-Storey frame-NF records-Capacity Curves RRSMV1

14000

12000

10000 Base Shear [kN]

8000

6000

4000

Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

2000

0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Top Floor Displacement [m]

Fig. 95. 9-Storey frame-NF records-Capacity Curves SCHMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

138

E1-ERZMV1

E1-LUCMV1

E1
70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% Uniform Triangular 1st Mode Code DAP

70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% Uniform Triangular 1st Mode Code DAP

Mean Max

Mean Max

E1-RRSMV1
300.00%

E1-SCHMV1

40.00% 35.00% 30.00% 25.00%

250.00%

200.00% Mean Max

150.00%

20.00% 15.00% 10.00% 5.00%

Mean Max

100.00%

50.00%

0.00% Uniform Triangular 1st Mode Code DAP 0.00% Uniform Triangular 1st Mode Code DAP

Fig. 96. 9-Storey frame-NF records-E1

E2-ERZMV1
1.80 1.60 1.40 1.20
Triangular 1st Mode Code DAP Uniform

E2-LUCMV1

1.80 1.60 1.40 1.20 1.00 0.80 0.60 0.40 0.20 0.00 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Top Floor TH displacement [m]

1.00 0.80 0.60 0.40 0.20 0.00 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Top Floor TH displacement [m]

Triangular 1st Mode Code DAP Uniform

E2-RRSMV1
1.40 1.20 1.00 Triangular 1st Mode Code DAP Uniform 0.80 0.60 0.40 0.20 0.50 0.00 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Top Floor TH displacement [m] 0.00 0 0.2 0.4 0.6

E2-SCHMV1

4.00 3.50 3.00 2.50 2.00 1.50 1.00

Triangular 1st Mode Code DAP Uniform

0.8

1.2

1.4

1.6

Top Floor TH displacement [m]

Fig. 97. 9-Storey frame-NF records-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

139

C.2.4 9-Storey weak frame

9000 8000 7000 6000 Base Shear [kN] 5000 4000 3000 2000 1000 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Top Floor Displacement [m] Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

Fig. 98. 9-Storey weak frame-NF records-Capacity Curves ERZMV1

8000 7000 6000 5000 4000 3000 2000 1000 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Top Floor Displacement [m] Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

Base Shear [kN]

Fig. 99. 9-Storey weak frame-NF records-Capacity Curves LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

140

9000 8000 7000 6000 Base Shear [kN] 5000 4000 3000 2000 1000 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Top Floor Displacement [m] Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

Fig. 100. 9-Storey weak frame-NF records-Capacity Curves RRSMV1

9000 8000 7000 6000 Base Shear [kN] 5000 4000 3000 2000 1000 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Top Floor Displacement [m] Max Disp. Vs. Corr. Base Shear Max Disp. vs. Max Base Shear corr. Disp. vs. Max Base Shear DAP Uniform Triangular 1st Mode Code

Fig. 101. 9-Storey weak frame-NF records-Capacity Curves SCHMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

141

E1-ERZMV1

E1-LUCMV1

40.00% 35.00% 30.00%

70.00% 60.00% 50.00%

25.00% 20.00% 15.00% 10.00% 5.00% 0.00% Uniform Triangular 1st Mode Code DAP Mean Max 40.00% 30.00% 20.00% 10.00% 0.00% Uniform Triangular 1st Mode Code DAP Mean Max

E1-RRSMV1
40.00% 35.00%

E1-SCHMV1

140.00% 120.00%

30.00%

100.00%
25.00%

80.00% 60.00% 40.00% 20.00% 0.00% Uniform Triangular 1st Mode Code DAP

Mean Max

20.00% 15.00% 10.00% 5.00% 0.00% Uniform Triangular 1st Mode Code DAP

Mean Max

Fig. 102. 9-Storey-weak frame-NF records-E1

E2-ERZMV1

E2-LUCMV1

1.60 1.40

3.00

2.50 1.20 1.00 0.80 0.60 0.40 0.20 0.00 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Top Floor TH displacement [m] 0.50 Triangular 1st Mode Code DAP Uniform 2.00 Triangular 1st Mode Code DAP Uniform

1.50

1.00

0.00 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Top Floor TH displacement [m]

E2-RRSMV1
2.50

E2-SCHMV1
1.40 1.20 1.00

2.00

1.50

1.00

Triangular 1st Mode Code DAP Uniform

0.80 0.60 0.40

Triangular 1st Mode Code DAP Uniform

0.50

0.20
0.00 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Top Floor TH displacement [m]

0.00 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Top Floor TH displacement [m]

Fig. 103. 9-Storey-weak frame-NF records-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

142

Appendix D: First parametric study - Storey response parameters

D.1 OGMs

D.1.1 3-Storey frame

Drift Level: 0.5% Drift Level: 0.5%

Storey 0.02 0.04 0.06 0.08 Floor Displacement [m] 0.1

Floor

0.2

0.4 0.6 0.8 Interstorey Drift [%]

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Evaluation of Pushover Procedures for the Seismic Design of Buildings

144

Drift Level: 0.5%

Drift Level: 0.5%

Storey

Storey 1000 2000 3000 Interstorey Shear [kN] 4000

0.5 1 1.5 2 Interstorey Moment [kNm]

2.5 x 10
4

Fig. 104. 3-Storey frame-total drift level 0.5%

Drift Level: 2%

Drift Level: 2%

Storey 0.1 0.2 Floor Displacement [m] 0.3

Floor

1 2 Interstorey Drift [%]

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Evaluation of Pushover Procedures for the Seismic Design of Buildings

145

Drift Level: 2%

Drift Level: 2%

Storey

Storey 2000 4000 6000 Interstorey Shear [kN]

1 2 3 4 Interstorey Moment [kNm]

5 x 10
4

Fig. 105. 3-Storey frame-total drift level 2%

Drift Level: 4%

Drift Level: 4%

Storey 0.2 0.4 Floor Displacement [m] 0.6

Floor

2 4 Interstorey Drift [%]

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Evaluation of Pushover Procedures for the Seismic Design of Buildings

146

Drift Level: 4%

Drift Level: 4%

Storey

Storey 1 2 3 4 Interstorey Moment [kNm] 5 x 10

4

1 2 3 4 Interstorey Moment [kNm]

5 x 10
4

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Fig. 106. 3-Storey frame-total drift level 4%

Evaluation of Pushover Procedures for the Seismic Design of Buildings

147

D.1.2 3-Storey weak frame

Drift Level: 0.5% Drift Level: 0.5%

Storey 0.02 0.04 0.06 Floor Displacement [m] Drift Level: 0.5% 0.08

Floor

0.2

0.4 0.6 0.8 Interstorey Drift [%] Drift Level: 0.5%

Storey

Storey 1000 2000 3000 Interstorey Shear [kN] 4000

1 2 Interstorey Moment [kNm]

3 x 10
4

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Fig. 107. 3-Storey-weak frame-total drift level 0.5%

Evaluation of Pushover Procedures for the Seismic Design of Buildings

148

Drift Level: 2%

Drift Level: 2%

Storey 0.2 0.4 Floor Displacement [m] Drift Level: 2% 0.6

Floor

2 3 4 Interstorey Drift [%] Drift Level: 2%

Storey

Storey 1000 2000 3000 4000 5000 Interstorey Shear [kN]

1 2 3 4 4 Interstorey Moment [kNm] x 10

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Fig. 108. 3-Storey-weak frame-total drift level 2%

Evaluation of Pushover Procedures for the Seismic Design of Buildings

149

Drift Level: 4%

Drift Level: 4%

Storey 0.2 0.4 Floor Displacement [m] Drift Level: 4% 0.6

Floor

4 6 8 Interstorey Drift [%] Drift Level: 4%

10

Storey

Storey 2000 4000 Interstorey Shear [kN] 6000

1 2 3 4 4 Interstorey Moment [kNm] x 10

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Fig. 109. 3-Storey-weak frame-total drift level 4%

Evaluation of Pushover Procedures for the Seismic Design of Buildings

150

D.1.3 9-Storey frame

Drift Level: 0.5% 10 9 8 7 6
Storey 10 9 8 7 6 5 4 3 2 1 0 0.2 Drift Level: 0.5%

Floor

5 4 3 2 1 0

0.05 0.1 0.15 Floor Displacement [m]

Drift Level: 0.5%

0.2

0.4

0.6 0.8 1 Interstorey Drift [%]

1.2

Drift Level: 0.5% 10 9 8 7 6 Storey 5 4 3 2 1 0

10 9 8 7 6 Storey 5 4 3 2 1 0

2000 4000 Interstorey Shear [kN]

6000

5 10 Interstorey Moment [kNm]

15 x 10
4

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Fig. 110. 9-Storey frame-total drift level 0.5%

Evaluation of Pushover Procedures for the Seismic Design of Buildings

151

Drift Level: 2% 10 9 8 7 6 5 4 3 2 1 0

Drift Level: 2% 10 9 8 7 6 Storey 5 4 3 2 1 0

Floor

0.2 0.4 0.6 Floor Displacement [m]

0.8

2 3 4 Interstorey Drift [%] Drift Level: 2%

Drift Level: 2% 10 9 8 7 6 Storey 5 4 3 2 1 0 Storey 0 5000 10000 Interstorey Shear [kN] 15000 10 9 8 7 6 5 4 3 2 1 0

0.5 1 1.5 2 Interstorey Moment [kNm]

2.5 x 10
5

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Fig. 111. 9-Storey frame-total drift level 2%

Evaluation of Pushover Procedures for the Seismic Design of Buildings

152

Drift Level: 4% 10 9 8 7 6 Storey Floor 5 4 3 2 1 0 10 9 8 7 6 5 4 3 2 1 0

Drift Level: 4%

0.5 1 1.5 Floor Displacement [m] Drift Level: 4%

5 10 Interstorey Drift [%] Drift Level: 4%

15

10 9 8 7 6 Storey 5 4 3 2 1 0 Storey 0 5000 10000 Interstorey Shear [kN] 15000

10 9 8 7 6 5 4 3 2 1 0

0.5 1 1.5 2 Interstorey Moment [kNm]

2.5 x 10
5

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Fig. 112. 9-Storey frame-total drift level 4%

Evaluation of Pushover Procedures for the Seismic Design of Buildings

153

D.1.4 9-Storey weak frame

Drift Level: 0.5% 10 9 8 7 6 5 4 3 2 1 0

Drift Level: 0.5% 10 9 8 7 6 Storey 5 4 3 2 1 0 0.2

Floor

0.05 0.1 0.15 Floor Displacement [m]

0.2

0.4

0.6 0.8 1 Interstorey Drift [%]

1.2

Drift Level: 0.5% 10 9 8 7 6 5 4 3 2 1 0 Storey Storey 10 9 8 7 6 5 4 3 2 1 0

Drift Level: 0.5%

2000 4000 Interstorey Shear [kN]

6000

0.5 1 1.5 Interstorey Moment [kNm]

2 x 10
5

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Fig. 113. 9-Storey weak frame-total drift level 0.5%

Evaluation of Pushover Procedures for the Seismic Design of Buildings

154

Drift Level: 2% 10 9 8 7 6
Storey 10 9 8 7 6 5 4 3 2 1 0

Drift Level: 2%

Floor

5 4 3 2 1 0

0.2 0.4 0.6 Floor Displacement [m]

Drift Level: 2%

0.8

4 6 8 Interstorey Drift [%] Drift Level: 2%

10

10 9 8 7 6 Storey 5 4 3 2 1 0 Storey 0 2000 4000 6000 8000 Interstorey Shear [kN] 10000

10 9 8 7 6 5 4 3 2 1 0

0.5 1 1.5 2 Interstorey Moment [kNm]

2.5 x 10
5

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Fig. 114. 9-Storey weak frame-total drift level 2%

Evaluation of Pushover Procedures for the Seismic Design of Buildings

155

Drift Level: 2.7% 10 9 8 7 6

Storey 10 9 8 7 6 5 4 3 2 1 0

Drift Level: 2.7%

Floor

5 4 3 2 1 0 0.2

0.4 0.6 0.8 1 Floor Displacement [m]

1.2

5 10 15 Interstorey Drift [%]

20

Drift Level: 2.7% 10 9 8 7 6 Storey 5 4 3 2 1 0 Storey 10 9 8 7 6 5 4 3 2 1 0

Drift Level: 2.7%

5000 10000 Interstorey Shear [kN]

15000

0.5 1 1.5 2 Interstorey Moment [kNm]

2.5 x 10
5

TH
Median Min SD SD Mean Max Min

DAP
Mean Median SD SD Max

Uniform
1st Mode

Triangular
Code

Average Spectrum

Fig. 115. 9-Storey weak frame-total drift level 2.7%

Evaluation of Pushover Procedures for the Seismic Design of Buildings

156

D.2
Record

NF records Table 19. Total drift recorded in the analyse [%]

ERZMV1 2.58 RRSMV1 1.95 LUCMV1 1.29 SCHMV1 1.78

RCW
3S

3Sw 9S 9Sw

4.18 3.61 1.677 1.64

5.0246 3.79 1.872 1.925

2.0947 2.95 1.829 2.33

3.0497 2.57 1.776 2.02

Evaluation of Pushover Procedures for the Seismic Design of Buildings

157

D.2.1 3-Storey frame

Storey 0.2 0.4 Floor Displacement [m] 0.6

Floor

2 4 Interstorey Drift [%]

Storey

Storey

2000 4000 6000 Interstorey Shear [kN]

1 2 3 4 Interstorey Moment [kNm]

5 x 10
4

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 116. 3 Storey frame-ERZMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

158

Storey

Floor

0.1 0.2 Floor Displacement [m]

0.3

1 2 Interstorey Drift [%]

Storey

Storey 2000 4000 6000 Interstorey Shear [kN]

1 2 3 4 Interstorey Moment [kNm]

5 x 10
4

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 117. 3 Storey frame-LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

159

Storey 0.2 0.4 0.6 Floor Displacement [m]

Floor

2 4 Interstorey Drift [%]

Storey

Storey

2000 4000 6000 Interstorey Shear [kN]

2 4 Interstorey Moment [kNm]

6 x 10
4

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 118. 3 Storey frame-RRSMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

160

Storey
0.1 0.2 0.3 0.4 Floor Displacement [m] 0.5

Floor

1 2 3 Interstorey Drift [%]

Storey

Storey 2000 4000 6000 Interstorey Shear [kN] 8000

2 4 Interstorey Moment [kNm]

6 x 10
4

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 119. 3 Storey frame-SCHMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

161

D.2.2 3-Storey weak frame

Storey
0.1 0.2 0.3 0.4 Floor Displacement [m] 0.5

Floor

4 6 8 Interstorey Drift [%]

10

Storey

Storey
2000 4000 Interstorey Shear [kN] 6000

1 2 3 4 Interstorey Moment [kNm]

5 x 10
4

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 120. 3-Storey weak-frame-ERZMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

162

Storey

Floor

0.1 0.2 0.3 0.4 Floor Displacement [m]

0.5

2 4 6 Interstorey Drift [%]

Storey

Storey 1000 2000 3000 4000 Interstorey Shear [kN] 5000

1 2 3 4 Interstorey Moment [kNm]

5 x 10
4

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 121. 3-Storey-weak frame-LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

163

Storey

Floor

0.2 0.4 Floor Displacement [m]

0.6

4 6 8 Interstorey Drift [%]

10

Storey

Storey 2000 4000 Interstorey Shear [kN] 6000

1 2 3 4 Interstorey Moment [kNm]

5 x 10
4

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 122. 3-Storey-weak frame-RRSMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

164

Storey 0.1 0.2 0.3 Floor Displacement [m] 0.4

Floor

2 4 Interstorey Drift [%]

Storey

Storey 1000 2000 3000 4000 Interstorey Shear [kN] 5000

1 2 3 4 Interstorey Moment [kNm]

5 x 10
4

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 123. 3-Storey-weak frame-SCHMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

165

D.2.3 9-Storey frame

10
10

6
Storey

Floor

0.1

0.2

0.3 0.4 0.5 Floor Displacement [m]

0.6

0.5

1.5 2 2.5 Interstorey Drift [%]

10

10

6 Storey
Storey

2000 3000 4000 5000 6000 7000 8000 9000 Interstorey Shear [kN]

0.5

1 1.5 Interstorey Moment [kNm]

2 x 10
5

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 124. 9-Storey -frame-ERZMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

166

10

10

6 Storey
0.1 0.2 0.3 0.4 0.5 Floor Displacement [m] 0.6

Floor

0.5

1.5 2 2.5 Interstorey Drift [%]

3.5

10

10
9

9
8

8
7

7
6

6
Storey

Storey

4
3

3
2

2
1

1
0 0.5 1 1.5 Interstorey Moment [kNm] 2 x 10
5

2000 3000 4000 5000 6000 7000 8000 9000 Interstorey Shear [kN]

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 125. 9-Storey -frame-LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

167

10

10

6 Storey
0.2 0.3 0.4 0.5 Floor Displacement [m] 0.6

Floor

0 0.1

0.5

1.5 2 2.5 Interstorey Drift [%]

3.5

10

10

6 Storey
Storey

2000

4000 6000 8000 Interstorey Shear [kN]

10000

0.5

1 1.5 Interstorey Moment [kNm]

2 x 10
5

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 126. 9-Storey -frame-RRSMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

168

10

10

6 Storey
0.1 0.2 0.3 0.4 0.5 Floor Displacement [m] 0.6

Floor

0.5

1.5 2 2.5 Interstorey Drift [%]

10

10

6 Storey
Storey

2000 3000 4000 5000 6000 7000 8000 9000 Interstorey Shear [kN]

0.5

1 1.5 Interstorey Moment [kNm]

2 x 10
5

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 127. 9-Storey -frame-SCHMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

169

D.2.4 9-Storey weak frame

10

10

6 Floor

6 Storey
0.2 0.3 0.4 0.5 Floor Displacement [m] 0.6

3 4 5 Interstorey Drift [%]

10

10

6 Storey
Storey

2000

4000 6000 Interstorey Shear [kN]

8000

0.5 1 1.5 Interstorey Moment [kNm]

2 x 10
5

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 128. 9-Storey weak-frame-ERZMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

170

10 9 8 7 6 Storey 5 4 3 2 1 0
Storey

10 9 8 7 6 5 4 3 2 1

0.5 1 1.5 Interstorey Moment [kNm]

2 x 10
5

4 6 8 10 Interstorey Drift [%]

12

10 9 8 7 6 Storey 5 4 3 2 1 0
Storey

10 9 8 7 6 5 4 3 2 1 0

2000

4000 6000 8000 Interstorey Shear [kN]

10000

0.5 1 1.5 Interstorey Moment [kNm]

2 x 10
5

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 129. 9-Storey weak-frame-LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

171

10

10

6 Storey
0.2 0.3 0.4 0.5 Floor Displacement [m] 0.6 0.7

Floor

4 6 Interstorey Drift [%]

10
10

9
9

6 Storey

6 Storey

0
2000 4000 6000 Interstorey Shear [kN] 8000

0.5 1 1.5 Interstorey Moment [kNm]

2 x 10
5

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 130. 9-Storey weak-frame-RRSMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

172

10

10

6 Floor

6 Storey
0.2 0.3 0.4 0.5 0.6 Floor Displacement [m] 0.7

4 6 Interstorey Drift [%]

10

10

6 Storey
Storey

2000

4000 6000 Interstorey Shear [kN]

8000

0.5 1 1.5 Interstorey Moment [kNm]

2 x 10
5

TH
Uniform
1st Mode

DAP
Triangular
Code

Fig. 131. 9-Storey weak-frame-SCHMV1

Appendix E: First parametric study - error measurement (OGMs)

E.1 Mean and Maximum E1

E.1.1 3-Storey frame

E1 - Maximum values 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS E1 - Maximum values

E1: maximum values drift 0.5%

E1 - Maximum values 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS

E1: maximum values drift 2%

E1: maximum values drift 4% Fig. 132. 3-Storey frame-E1 (maximum values)

Evaluation of Pushover Procedures for the Seismic Design of Buildings

174

E1 - Mean values 35% 30% 25% 20% 15% 10% 5% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS 35% 30% 25% 20% 15% 10% 5% 0% Floor Disp.

E1 - Mean values

DAP Uniform Triangular 1st Mode Code DAP - AS

Inter-storey Drift Inter-storey Shear

Inter-storey Moment

E1: mean values drift 0.5%

E1 - Mean values 35% 30% 25% 20% 15% 10% 5% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS

E1: mean values drift 2%

E1: mean values drift 4% Fig. 133. 3-Storey frame-E1 (mean values) Table 20. 3-Storey frame-drift level 0.5%-E1 (mean)
Mean E1 Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment DAP
8.88% 15.79% 30.67% 23.16%

Uniform
0.99% 14.34% 28.78% 27.21%

Triangular
8.34% 15.70% 30.51% 23.60%

1st Mode
9.84% 15.95% 30.83% 22.89%

Code
9.84% 15.95% 30.83% 22.89%

DAP - AS
8.84% 15.78% 30.65% 23.21%

Table 21. 3-Storey frame-drift level 0.5%-E1 (standard deviation)

St.Dev. E1 DAP
9.16% 6.11% 11.60% 18.26%

Uniform
0.42% 17.06% 22.63% 26.63%

Triangular
9.11% 7.90% 13.71% 19.51%

1st Mode
10.54% 7.73% 12.43% 18.05%

Code
10.54% 7.73% 12.43% 18.05%

DAP - AS
9.20% 6.39% 11.90% 18.42%

Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment

Evaluation of Pushover Procedures for the Seismic Design of Buildings

175

Table 22. 3-Storey frame-drift level 2%-E1 (mean)

Mean E1 Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment DAP
8.82% 8.66% 28.14% 17.94%

Uniform
10.52% 20.27% 29.09% 26.34%

Triangular
8.51% 8.70% 28.29% 18.31%

1st Mode
10.00% 8.90% 28.30% 17.69%

Code
12.02% 10.66% 28.31% 16.87%

DAP - AS
8.66% 8.67% 28.23% 18.10%

Table 23. 3-Storey frame-drift level 2%-E1 (standard deviation)

St.Dev. E1 DAP
8.77% 8.36% 10.59% 19.79%

Uniform
10.50% 13.82% 24.78% 26.52%

Triangular
9.11% 8.30% 11.58% 21.01%

1st Mode
10.71% 10.93% 10.72% 20.74%

Code
12.60% 12.73% 9.07% 19.80%

DAP - AS
8.88% 8.19% 10.96% 20.31%

Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment

Table 24. 3-Storey frame-drift level 4%-E1 (mean)

Mean E1 Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment DAP
3.48% 5.33% 29.24% 18.51%

Uniform
9.21% 15.50% 30.52% 26.50%

Triangular
2.73% 5.22% 29.82% 19.42%

1st Mode
3.63% 5.36% 29.78% 18.92%

Code
4.90% 5.57% 29.70% 18.13%

DAP - AS
3.06% 5.27% 29.74% 19.01%

Table 25. 3-Storey frame-drift level 4%-E1 (standard deviation)

St.Dev. E1 DAP
2.75% 1.19% 8.82% 18.71%

Uniform
8.88% 11.57% 23.13% 27.19%

Triangular
2.86% 3.05% 11.40% 21.54%

1st Mode
3.82% 3.25% 10.73% 21.14%

Code
4.98% 4.21% 9.26% 20.17%

DAP - AS
2.70% 1.20% 9.91% 20.24%

Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment

E.1.2 3-Storey weak frame

E1 - Maximum values 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS E1 - Maximum values

E1: maximum values drift 0.5%

E1: maximum values drift 2%

Evaluation of Pushover Procedures for the Seismic Design of Buildings

176

E1 - Maximum values 80% 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS

E1: maximum values drift 4% Fig. 134. 3-Storey weak frame-E1 (maximum values)

E1 - Mean values 30% 25% 20% 15% 10% 5% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS 80% 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp.

E1 - Mean values

DAP Uniform Triangular 1st Mode Code DAP - AS

Inter-storey Drift Inter-storey Shear

Inter-storey Moment

E1: mean values drift 0.5%

E1 - Mean values 120% 100% 80% 60% 40% 20% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS

E1: mean values drift 2%

E1: mean values drift 4% Fig. 135. 3-Storey weak frame-E1 (mean values) Table 26. 3-Storey weak frame-drift level 0.5%-E1 (mean)
Mean E1 Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment DAP
6.94% 12.32% 26.74% 19.91%

Uniform
4.81% 13.45% 24.89% 26.40%

Triangular
6.36% 12.22% 26.56% 20.38%

1st Mode
7.03% 12.33% 26.72% 20.11%

Code
7.94% 12.49% 26.95% 19.64%

DAP - AS
6.90% 12.32% 26.72% 19.96%

Evaluation of Pushover Procedures for the Seismic Design of Buildings

177

Table 27. 3-Storey weak frame-drift level 0.5%-E1 (standard deviation)

St.Dev. E1 DAP
7.06% 6.96% 12.51% 19.40%

Uniform
1.95% 15.99% 25.04% 25.31%

Triangular
7.32% 8.88% 14.66% 20.68%

1st Mode
8.00% 9.06% 14.34% 20.22%

Code
8.54% 8.55% 13.03% 19.16%

DAP - AS
7.13% 7.23% 12.79% 19.55%

Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment

Table 28. 3-Storey weak frame-drift level 2%-E1 (mean)

Mean E1 Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment DAP
4.91% 23.40% 23.67% 22.72%

Uniform
14.08% 43.60% 35.32% 38.26%

Triangular
4.33% 21.87% 23.14% 22.45%

1st Mode
2.60% 17.99% 21.81% 20.25%

Code
0.61% 13.80% 20.45% 19.26%

DAP - AS
5.19% 23.91% 23.88% 23.11%

Table 29. 3-Storey weak frame-drift level 2%-E1 (standard deviation)

St.Dev. E1 DAP
5.56% 11.33% 20.48% 19.56%

Uniform
15.89% 12.01% 27.78% 23.67%

Triangular
4.70% 11.80% 20.84% 20.06%

1st Mode
2.70% 11.80% 20.46% 15.58%

Code
0.61% 11.33% 19.60% 18.90%

DAP - AS
5.79% 11.54% 20.81% 19.90%

Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment

Table 30. 3-Storey weak frame-drift level 4%-E1 (mean)

Mean E1 Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment DAP
2.37% 23.16% 27.08% 24.65%

Uniform
10.60% 47.59% 39.27% 40.19%

Triangular
2.97% 24.66% 27.52% 25.43%

1st Mode
1.08% 19.12% 26.04% 24.40%

Code
1.66% 14.43% 24.97% 23.15%

DAP - AS
2.97% 24.83% 27.53% 25.21%

Table 31. 3-Storey weak frame-drift level 4%-E1 (standard deviation)

St.Dev. E1 DAP
2.11% 18.08% 23.80% 24.46%

Uniform
11.11% 23.44% 29.92% 28.39%

Triangular
2.63% 19.00% 24.39% 25.21%

1st Mode
0.62% 17.72% 24.27% 24.31%

Code
1.60% 12.07% 23.07% 22.85%

DAP - AS
2.71% 18.70% 24.05% 24.84%

Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment

Evaluation of Pushover Procedures for the Seismic Design of Buildings

178

E.1.3 9-Storey frame

E1 - Maximum values 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS 100% 80% 60% 40% 20% DAP Uniform Triangular 1st Mode Code DAP - AS E1 - Maximum values

E1: maximum values drift 0.5%

E1 - Maximum values 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS

E1: maximum values drift 2%

E1: maximum values drift 4% Fig. 136. 9-Storey frame-E1 (maximum values)

E1 - Mean values 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment 0% Floor Disp. DAP Uniform Triangular 1st Mode Code DAP - AS 100% 80% 60% 40% 20%

E1 - Mean values

DAP Unifo Trian 1st M Code DAP

Inter-storey Drift Inter-storey Shear

Inter-storey Moment

E1: mean values drift 0.5%

E1: mean values drift 2%

Evaluation of Pushover Procedures for the Seismic Design of Buildings

179

E1 - Mean values 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS

E1: mean values drift 4% Fig. 137. 9-Storey frame-E1 (mean values) Table 32. 9-Storey frame-drift level 0.5%-E1 (mean)
Mean E1 Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment DAP
3.20% 18.20% 29.42% 27.35%

Uniform
13.64% 26.94% 32.53% 37.70%

Triangular
4.07% 17.81% 28.40% 29.73%

1st Mode
4.42% 17.81% 28.24% 30.49%

Code
6.97% 17.78% 29.22% 25.16%

DAP - AS
3.18% 18.19% 29.41% 27.39%

Table 33. 9-Storey frame-drift level 0.5%-E1 (standard deviation)

St.Dev. E1 DAP
3.97% 8.97% 16.95% 20.94%

Uniform
6.74% 18.37% 25.90% 25.98%

Triangular
2.63% 14.87% 21.21% 23.86%

1st Mode
2.17% 16.24% 22.26% 24.32%

Code
7.33% 8.12% 15.68% 21.66%

DAP - AS
3.96% 9.04% 17.00% 20.97%

Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment

Table 34. 9-Storey frame-drift level 2%-E1 (mean)

Mean E1 Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment DAP
6.17% 22.82% 28.86% 20.29%

Uniform
32.77% 55.72% 39.20% 43.98%

Triangular
10.19% 31.10% 31.80% 29.67%

1st Mode
12.07% 33.25% 32.40% 31.03%

Code
7.43% 23.12% 29.82% 22.74%

DAP - AS
3.36% 23.19% 28.66% 20.67%

Table 35. 9-Storey frame-drift level 2%-E1 (standard deviation)

St.Dev. E1 DAP
3.37% 8.32% 14.33% 19.10%

Uniform
28.04% 26.57% 29.39% 26.83%

Triangular
5.04% 27.37% 22.47% 26.14%

1st Mode
6.24% 27.20% 23.30% 26.57%

Code
7.32% 18.52% 17.02% 23.30%

DAP - AS
2.17% 11.54% 14.89% 19.41%

Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment

Evaluation of Pushover Procedures for the Seismic Design of Buildings

180

Table 36. 9-Storey frame-drift level 4%-E1 (mean)

Mean E1 Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment St.Dev. E1 DAP
22.92% 26.16% 35.71% 20.14%

Uniform
18.11% 50.87% 44.42% 46.62%

Triangular
5.34% 31.21% 38.22% 33.15%

1st Mode
7.02% 33.97% 38.72% 34.46%

Code
12.87% 18.26% 35.38% 24.62%

DAP - AS
22.80% 26.74% 35.90% 20.15%

Table 37. 9-Storey frame-drift level 4%-E1 (standard deviation)

DAP
12.25% 9.28% 11.49% 19.74%

Uniform
14.37% 30.46% 29.26% 27.52%

Triangular
3.03% 31.13% 22.66% 27.23%

1st Mode
4.09% 31.32% 23.45% 27.55%

Code
12.05% 18.47% 17.14% 25.02%

DAP - AS
12.35% 8.30% 12.12% 20.39%

Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment

E.1.4 9-Storey weak frame

E1 - Maximum values 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS E1 - Maximum values

E1: maximum values drift 0.5%

E1 - Maximum values 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS

E1: maximum values drift 2%

E1: maximum values drift 2.7% Fig. 138. 9-Storey weak frame-E1 (maximum values)

Evaluation of Pushover Procedures for the Seismic Design of Buildings

181

E1 - Mean values 80% 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS

E1 - Mean values 80% 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangula 1st Mode Code DAP - AS

E1: mean values drift 0.5%

E1 - Mean values 80% 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS

E1: mean values drift 2%

E1: mean values drift 2.7% Fig. 139. 9-Storey weak frame-E1 (mean values) Table 38. 9-Storey weak frame-drift level 0.5%-E1 (mean)
Mean E1 Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment DAP
3.09% 18.07% 25.17% 21.42%

Uniform
14.10% 26.80% 29.54% 47.85%

Triangular
3.77% 17.56% 24.30% 48.38%

1st Mode
4.12% 17.66% 24.19% 48.35%

Code
6.79% 17.63% 24.81% 48.65%

DAP - AS
2.91% 18.04% 25.12% 21.63%

Table 39. 9-Storey weak frame-drift level 0.5%-E1 (standard deviation)

St.Dev. E1 DAP
3.36% 9.66% 14.21% 17.91%

Uniform
7.22% 19.37% 24.05% 23.94%

Triangular
2.20% 15.87% 19.30% 23.82%

1st Mode
2.05% 17.08% 20.42% 23.84%

Code
6.65% 9.27% 13.68% 23.72%

DAP - AS
3.29% 10.10% 14.56% 18.15%

Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment

Evaluation of Pushover Procedures for the Seismic Design of Buildings

182

Table 40. 9-Storey weak frame-drift level 2%-E1 (mean)

Mean E1 Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment DAP
3.11% 28.88% 25.82% 16.86%

Uniform
34.62% 73.88% 46.68% 19.21%

Triangular
13.98% 42.73% 32.84% 17.68%

1st Mode
15.93% 45.38% 33.95% 17.81%

Code
5.71% 28.69% 26.15% 16.73%

DAP - AS
2.70% 29.08% 25.25% 15.96%

Table 41. 9-Storey weak frame-drift level 2%-E1 (standard deviation)

St.Dev. E1 DAP
4.10% 17.86% 14.31% 16.39%

Uniform
38.38% 20.73% 22.72% 16.84%

Triangular
11.18% 25.26% 21.23% 16.86%

1st Mode
13.45% 24.50% 21.68% 16.90%

Code
7.54% 24.52% 16.90% 16.66%

DAP - AS
4.03% 18.49% 14.24% 16.66%

Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment

Table 42. 9-Storey weak frame-drift level 2.7%-E1 (mean)

Mean E1 Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment DAP
12.29% 27.10% 27.23% 14.78%

Uniform
21.83% 70.84% 47.64% 17.39%

Triangular
10.91% 47.49% 35.19% 15.91%

1st Mode
11.99% 49.51% 36.07% 16.04%

Code
5.98% 29.73% 29.08% 15.05%

DAP - AS
35.03% 40.00% 29.78% 17.04%

Table 43. 9-Storey weak frame-drift level 2.7%-E1 (standard deviation)

St.Dev. E1 DAP
14.00% 13.60% 12.58% 16.36%

Uniform
19.58% 10.71% 24.41% 16.93%

Triangular
6.82% 22.20% 21.59% 16.97%

1st Mode
7.90% 21.56% 22.31% 17.00%

Code
7.04% 25.51% 16.71% 16.74%

DAP - AS
10.91% 16.24% 12.30% 17.66%

Floor Disp. Interstorey Drift Interstorey Shear Interstorey Moment

E.2

Values of E1 and E2 for each building model

E.2.1 3-Storey frame

E1 - Floor Displacement 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey E1 - Inter-storey Drift

Evaluation of Pushover Procedures for the Seismic Design of Buildings

183

E1 - Inter-storey Shear 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

DAP - AS

Fig. 140. 3-Storey frame-total drift level 0.5%-E1

E2 - Floor Displacement 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 E2 - Inter-storey Drift 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP - AS

DAP

Uniform

Triangular

1st Mode

Code

DAP - AS

E2 - Inter-storey Shear 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

E2 - Moment 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP - AS

DAP

Uniform

Triangular

1st Mode

Code

DAP - AS

Fig. 141. 3-Storey frame-total drift level 0.5%-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

184

E1 - Floor Displacement 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey

Triangular

1st Mode E1 - Moment

Code

DAP - AS

E1 - Inter-storey Shear 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 60% 50% 40% 30% 20% 10% 0% DAP Uniform

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

DAP - AS

Fig. 142. 3-Storey frame-total drift level 2%-E1

E2 - Floor Displacement 1st storey 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 2nd storey 3rd storey 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular E2 - Moment 3rd storey 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 1st Mode Code DAP - AS E2 - Inter-storey Drift 1st storey 2nd storey 3rd storey

E2 - Inter-storey Shear 1st storey 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey

Fig. 143. 3-Storey frame-total drift level 2%-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

185

E1 - Floor Displacement 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey

Triangular

1st Mode E1 - Moment

Code

DAP - AS

E1 - Inter-storey Shear 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 60% 50% 40% 30% 20% 10% 0% DAP Uniform

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

DAP - AS

Fig. 144. 3-Storey frame-total drift level 4%-E1

E2 - Floor Displacement 1st storey 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 2nd storey 3rd storey 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular E2 - Moment 3rd storey 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 1st Mode Code DAP - AS E2 - Inter-storey Drift 1st storey 2nd storey 3rd storey

E2 - Inter-storey Shear 1st storey 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey

Fig. 145. 3-Storey frame-total drift level 4%-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

186

E.2.2 3-Storey weak frame

E1 - Floor Displacement 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey E1 - Inter-storey Drift

E1 - Inter-storey Shear 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

DAP - AS

Fig. 146. 3-Storey weak-frame-total drift level 0.5%-E1

E2 - Floor Displacement 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

E2 - Inter-storey Drift 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP - AS

DAP

Uniform

Triangular E2 - Moment

1st Mode

Code

DAP - AS

E2 - Inter-storey Shear 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform

1st storey

2nd storey

3rd storey

Triangular

1st Mode

Code

DAP - AS

Fig. 147. 3-Storey weak-frame-total drift level 0.5%-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

187

E1 - Floor Displacement 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

DAP - AS

E1 - Inter-storey Shear 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey

E1 - Moment 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey

Fig. 148. 3-Storey weak-frame-total drift level 2%-E1

E2 - Floor Displacement 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS E2 - Inter-storey Drift 1st storey 2nd storey 3rd storey

E2 - Inter-storey Shear 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform

E2 - Moment 1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

DAP - AS

Fig. 149. 3-Storey weak-frame-total drift level 2 %-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

188

E1 - Floor Displacement 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

DAP - AS

E1 - Inter-storey Shear 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

DAP - AS

Fig. 150. 3-Storey weak-frame-total drift level 4%-E1

E2 - Floor Displacement 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular E2 - Moment 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 1st Mode Code DAP - AS E2 - Inter-storey Drift 1st storey 2nd storey 3rd storey

E2 - Inter-storey Shear 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey

Fig. 151. 3-Storey weak-frame-total drift level 4%-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

189

E.2.3 9-Storey frame

E1 - Floor Displacement 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey
90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey E1 - Inter-storey Drift

E1 - Inter-storey Shear 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

DAP - AS

Fig. 152. 9-Storey frame-total drift level 0.5%-E1

E2 - Floor Displacement 1st storey 7th storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 0.6 0.4 0.2 0 DAP Uniform Triangular E2 - Moment 5th storey 6th storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 7th storey 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 1st Mode Code DAP - AS 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 1.4 1.2 1 0.8 1st storey 7th storey 2nd storey 8th storey E2 - Inter-storey Drift 3rd storey 9th storey 4th storey 5th storey 6th storey

E2 - Inter-storey Shear 1st storey 7th storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 8th storey 3rd storey 9th storey 4th storey

Fig. 153. 9-Storey frame-total drift level 0.5%-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

190

E1 - Floor Displacement

E1 - Inter-storey Drift

100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

E1 - Inter-storey Shear

E1 - Moment

100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Fig. 154. 9-Storey frame-total drift level 2%-E1

E2 - Floor Displacement 1st storey 7th storey 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular E2 - Moment 5th storey 6th storey 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 7th storey 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 1st Mode Code DAP - AS 1st storey 7th storey 2nd storey 8th storey E2 - Inter-storey Drift 3rd storey 9th storey 4th storey 5th storey 6th storey

E2 - Inter-storey Shear 1st storey 7th storey 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 8th storey 3rd storey 9th storey 4th storey

Fig. 155. 9-Storey frame-total drift level 2%-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

191

E1 - Floor Displacement 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

DAP - AS

E1 - Inter-storey Shear 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

DAP - AS

Fig. 156. 9-Storey frame-total drift level 4%-E1

E2 - Floor Displacement 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 DAP Uniform Triangular 1st Mode Code DAP - AS 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 DAP Uniform Triangular E2 - Moment 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 DAP Uniform Triangular 1st Mode Code DAP - AS DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 7th storey 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 1st Mode Code DAP - AS 1st storey 7th storey 2nd storey 8th storey E2 - Inter-storey Drift 3rd storey 9th storey 4th storey 5th storey 6th storey

E2 - Inter-storey Shear 1st storey 7th storey 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 2nd storey 8th storey 3rd storey 9th storey 4th storey

Fig. 157. 9-Storey frame-total drift level 4%-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

192

E.2.4 9-Storey weak frame

E1 - Floor Displacement 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey
90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey E1 - Inter-storey Drift

E1 - Inter-storey Shear 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

DAP - AS

Fig. 158. 9-Storey weak-frame-total drift level 0.5%-E1

E2 - Floor Displacement 1st storey 7th storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 1.4 1.2 1 0.8 1st storey 7th storey 2nd storey 8th storey E2 - Inter-storey Drift 3rd storey 9th storey 4th storey 5th storey 6th storey

E2 - Inter-storey Shear 1st storey 7th storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform 1st storey 7th storey 2nd storey 8th storey

E2 - Moment 3rd storey 9th storey 4th storey 5th storey 6th storey

Triangular

1st Mode

Code

DAP - AS

Fig. 159. 9-Storey weak-frame-total drift level 0.5%-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

193

E1 - Floor Displacement 120% 100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code DAP - AS
120% 100% 80% 60% 40% 20% 0% DAP Uniform

E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

DAP - AS

E1 - Inter-storey Shear 120% 100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 120% 100% 80% 60% 40% 20% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

DAP - AS

Fig. 160. 9-Storey weak-frame-total drift level 2%-E1

E2 - Floor Displacement 1st storey 7th storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 0 DAP Uniform Triangular 1st Mode Code DAP - AS 1 0.5 2 1.5 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 2.5 1st storey 7th storey 2nd storey 8th storey E2 - Inter-storey Drift 3rd storey 9th storey 4th storey 5th storey 6th storey

E2 - Inter-storey Shear 1st storey 7th storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform 1st storey 7th storey 2nd storey 8th storey

E2 - Moment 3rd storey 9th storey 4th storey 5th storey 6th storey

Triangular

1st Mode

Code

DAP - AS

Fig. 161. 9-Storey weak-frame-total drift level 2%-E2

Evaluation of Pushover Procedures for the Seismic Design of Buildings

194

E1 - Floor Displacement 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

E1 - Inter-storey Drift 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

E1 - Inter-storey Shear 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

DAP - AS

Fig. 162. 9-Storey weak-frame-total drift level 2.7%-E1

E2 - Floor Displacement 1st storey 7th storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular E2 - Moment 5th storey 6th storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 DAP Uniform Triangular 1st Mode Code DAP - AS DAP Uniform Triangular 1st Mode Code DAP - AS 1st storey 7th storey 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 1st Mode Code DAP - AS 1st storey 7th storey 2nd storey 8th storey E2 - Inter-storey Drift 3rd storey 9th storey 4th storey 5th storey 6th storey

E2 - Inter-storey Shear 1st storey 7th storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 8th storey 3rd storey 9th storey 4th storey

Fig. 163. 9-Storey weak-frame-total drift level 2.7%-E2

Appendix F: First parametric study - error measurement (NF records)

F.1 Mean and maximum E1
E1 - Mean values 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS E1 - Maximum values

Fig. 164. 3-Storey frame

E1 - Mean values 80% 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS 80% 70% 60% 50% 40% 30% 20% 10% 0% Floor Disp.

E1 - Maximum values

DAP Uniform Triangular 1st Mode Code DAP - AS

Inter-storey Drift Inter-storey Shear

Inter-storey Moment

Fig. 165. 3-Storey weak frame

Evaluation of Pushover Procedures for the Seismic Design of Buildings

196

E1 - Mean values 200% 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS 200% 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% Floor Disp.

E1 - Maximum values

DAP Uniform Triangular 1st Mode Code DAP - AS

Inter-storey Drift Inter-storey Shear

Inter-storey Moment

Fig. 166. 9-Storey frame

E1 - Mean values 200% 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% Floor Disp. Inter-storey Drift Inter-storey Shear Inter-storey Moment DAP Uniform Triangular 1st Mode Code DAP - AS 200% 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% Floor Disp.

E1 - Maximum values

DAP Uniform Triangular 1st Mode Code DAP - AS

Inter-storey Drift Inter-storey Shear

Inter-storey Moment

Fig. 167. 9-Storey weak frame

F.2

3-Storey frame
E1 - Floor Displacement E1 - Floor Displacement 40% 35% 30% 1st storey 2nd storey 3rd storey 25% 20% 15% 10% 5% 0% DAP Uniform Triangular 1st Mode Code DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey

40% 35% 30% 25% 20% 15% 10% 5% 0%

ERZMV1

LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

197

E1 - Floor Displacement 40% 35% 30% 25% 20% 15% 10% 5% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 40% 35% 30% 25% 20% 15% 10% 5% 0% DAP Uniform

E1 - Floor Displacement

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

RRSMV1

SCHMV1 Fig. 168. 3-Storey frame-E1-Floor Displacement

E1 - Inter-storey Drift 40% 35% 30% 25% 20% 15% 10% 5% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 40% 35% 30% 25% 20% 15% 10% 5% 0% DAP Uniform

E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

ERZMV1
E1 - Inter-storey Drift 40% 35% 30% 25% 20% 15% 10% 5% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 40% 35% 30% 25% 20% 15% 10% 5% 0% DAP Uniform

LUCMV1
E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

RRSMV1 Fig. 169. 3-Storey frame-E1-Interstorey Drift

SCHMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

198

E1 - Inter-storey Shear 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Inter-storey Shear

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

ERZMV1
E1 - Inter-storey Shear 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 60% 50% 40% 30% 20% 10% 0% DAP Uniform

LUCMV1
E1 - Inter-storey Shear

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

RRSMV1 Fig. 170. 3-Storey frame-E1-Interstorey Shear

SCHMV1

E1 - Moment 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

ERZMV1

LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

199

E1 - Moment 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

RRSMV1

SCHMV1 Fig. 171. 3-Storey frame-E1-Interstorey Moment

E2 - Floor Displacement 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

E2 - Floor Displacement 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1
E2 - Floor Displacement 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

LUCMV1
E2 - Floor Displacement 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1 Fig. 172. 3-Storey frame-E2-Floor Displacement

SCHMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

200

E2 - Inter-storey Drift 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

E2 - Inter-storey Drift 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1
E2 - Inter-storey Drift 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

LUCMV1
E2 - Inter-storey Drift 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1 Fig. 173. 3-Storey frame-E2-Interstorey Drift

SCHMV1

E2 - Inter-storey Shear 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

E2 - Inter-storey Shear 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1

LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

201

E2 - Inter-storey Shear 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

E2 - Inter-storey Shear 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1 Fig. 174. 3-Storey frame-E1-Interstorey Shear

SCHMV1

E2 - Moment 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

E2 - Moment 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1
E2 - Moment 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

LUCMV1
E2 - Moment 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1 Fig. 175. 3-Storey frame-E1-Interstorey Moment

SCHMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

202

F.3

3-Storey weak frame

E1 - Floor Displacement E1 - Floor Displacement 60% 50% 40% 1st storey 2nd storey 3rd storey 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey

60% 50% 40% 30% 20% 10% 0%

ERZMV1
E1 - Floor Displacement 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 60% 50% 40% 30% 20% 10% 0% DAP Uniform

LUCMV1
E1 - Floor Displacement

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 176. 3-Storey-weak frame-E1-Floor Displacement

E1 - Inter-storey Drift 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

ERZMV1

LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

203

E1 - Inter-storey Drift 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 177. 3-Storey-weak frame-E1-Interstorey Drift

E1 - Inter-storey Shear 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Inter-storey Shear

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

ERZMV1
E1 - Inter-storey Shear 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

LUCMV1
E1 - Inter-storey Shear

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 178. 3-Storey-weak frame-E1-Interstorey Shear

Evaluation of Pushover Procedures for the Seismic Design of Buildings

204

E1 - Moment 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

ERZMV1
E1 - Moment 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

LUCMV1
E1 - Moment

1st storey 2nd storey 3rd storey

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 179. 3-Storey-weak frame-E1-Interstorey Moment

E2 - Floor Displacement 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

E2 - Floor Displacement 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1

LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

205

E2 - Floor Displacement 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

E2 - Floor Displacement 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 180. 3-Storey-weak frame-E2-Floor Displacement

E2 - Inter-storey Drift 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

E2 - Inter-storey Drift 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1
E2 - Inter-storey Drift 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

LUCMV1
E2 - Inter-storey Drift 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 181. 3-Storey-weak frame-E2-Interstorey Drift

Evaluation of Pushover Procedures for the Seismic Design of Buildings

206

E2 - Inter-storey Shear 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

E2 - Inter-storey Shear 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1
E2 - Inter-storey Shear 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

LUCMV1
E2 - Inter-storey Shear 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 182. 3-Storey-weak frame-E1-Interstorey Shear

E2 - Moment 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

E2 - Moment 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1

LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

207

E2 - Moment 1st storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 2nd storey 3rd storey 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

E2 - Moment 1st storey 2nd storey 3rd storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 183. 3-Storey-weak frame-E1-Interstorey Moment

F.4

9-Storey frame
E1 - Floor Displacement E1 - Floor Displacement 120% 100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code DAP Uniform Triangular 1st Mode Code

120% 100% 80% 60% 40% 20% 0%

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

ERZMV1
E1 - Floor Displacement 120% 100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code 120% 100% 80% 60% 40% 20% 0% DAP Uniform

LUCMV1
E1 - Floor Displacement

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

RRSMV1

SCHMV1 Fig. 184. 9-Storey frame-E1-Floor Displacement

Evaluation of Pushover Procedures for the Seismic Design of Buildings

208

E1 - Inter-storey Drift 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

ERZMV1
E1 - Inter-storey Drift 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

LUCMV1
E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

RRSMV1 Fig. 185. 9-Storey frame-E1-Interstorey Drift

SCHMV1

E1 - Inter-storey Shear 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform

E1 - Inter-storey Shear

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

ERZMV1

LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

209

E1 - Inter-storey Shear 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform

E1 - Inter-storey Shear

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

RRSMV1 Fig. 186. 9-Storey frame-E1-Interstorey Shear

SCHMV1

E1 - Moment 200% 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 200% 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

ERZMV1
E1 - Moment 200% 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 200% 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform

LUCMV1
E1 - Moment

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

RRSMV1

SCHMV1 Fig. 187. 9-Storey frame-E1-Interstorey Moment

Evaluation of Pushover Procedures for the Seismic Design of Buildings

210

E2 - Floor Displacement 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey

E2 - Floor Displacement 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1
E2 - Floor Displacement 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey 2nd storey 8th storey

LUCMV1
E2 - Floor Displacement 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1 Fig. 188. 9-Storey frame-E2-Floor Displacement

SCHMV1

E2 - Inter-storey Drift 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey 2nd storey 8th storey

E2 - Inter-storey Drift 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1

LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

211

E2 - Inter-storey Drift 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey 2nd storey 8th storey

E2 - Inter-storey Drift 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1 Fig. 189. 9-Storey frame-E2-Interstorey Drift

SCHMV1

E2 - Inter-storey Shear 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey

E2 - Inter-storey Shear 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1
E2 - Inter-storey Shear 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey 2nd storey 8th storey

LUCMV1
E2 - Inter-storey Shear 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1 Fig. 190. 9-Storey frame-E1-Interstorey Shear

SCHMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

212

E2 - Moment 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey 2nd storey 8th storey

E2 - Moment 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1
E2 - Moment 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey

LUCMV1
E2 - Moment 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1 Fig. 191. 9-Storey frame-E1-Interstorey Moment

SCHMV1

F.5

9-Storey weak frame

E1 - Floor Displacement E1 - Floor Displacement 180% 160% 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

180% 160% 140% 120% 100% 80% 60% 40% 20% 0%

ERZMV1

LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

213

E1 - Floor Displacement 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform

E1 - Floor Displacement

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 192. 9-Storey-weak frame-E1-Floor Displacement

E1 - Inter-storey Drift 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform

E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

ERZMV1
E1 - Inter-storey Drift 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% DAP Uniform

LUCMV1
E1 - Inter-storey Drift

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 193. 9-Storey-weak frame-E1-Interstorey Drift

Evaluation of Pushover Procedures for the Seismic Design of Buildings

214

E1 - Inter-storey Shear 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Inter-storey Shear

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

ERZMV1
E1 - Inter-storey Shear 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

LUCMV1
E1 - Inter-storey Shear

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 194. 9-Storey-weak frame-E1-Interstorey Shear

E1 - Moment 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

ERZMV1

LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

215

E1 - Moment 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform Triangular 1st Mode Code 1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% DAP Uniform

E1 - Moment

1st storey 2nd storey 3rd storey 4th storey 5th storey 6th storey 7th storey 8th storey 9th storey

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 195. 9-Storey-weak frame-E1-Interstorey Moment

E2 - Floor Displacement 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey

E2 - Floor Displacement 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1
E2 - Floor Displacement 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey 2nd storey 8th storey

LUCMV1
E2 - Floor Displacement 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 196. 9-Storey-weak frame-E2-Floor Displacement

Evaluation of Pushover Procedures for the Seismic Design of Buildings

216

E2 - Inter-storey Drift 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey 2nd storey 8th storey

E2 - Inter-storey Drift 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1
E2 - Inter-storey Drift 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey 2nd storey 8th storey

LUCMV1
E2 - Inter-storey Drift 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 197. 9-Storey-weak frame-E2-Interstorey Drift

E2 - Inter-storey Shear 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey

E2 - Inter-storey Shear 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1

LUCMV1

Evaluation of Pushover Procedures for the Seismic Design of Buildings

217

E2 - Inter-storey Shear 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey

E2 - Inter-storey Shear 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 198. 9-Storey-weak frame-E1-Interstorey Shear

E2 - Moment 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey 2nd storey 8th storey

E2 - Moment 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

ERZMV1
E2 - Moment 1st storey 7th storey 3.5 3 2.5 2 1.5 1 0.5 0 2nd storey 8th storey 3rd storey 9th storey 4th storey 5th storey 6th storey 3.5 3 2.5 2 1.5 1 0.5 0 1st storey 7th storey 2nd storey 8th storey

LUCMV1
E2 - Moment 3rd storey 9th storey 4th storey 5th storey 6th storey

DAP

Uniform

Triangular

1st Mode

Code

DAP

Uniform

Triangular

1st Mode

Code

RRSMV1

SCHMV1

Fig. 199. 9-Storey-weak frame-E1-Interstorey Moment

Evaluation of Pushover Procedures for the Seismic Design of Buildings

218

Appendix G: Second parametric study

G.1 Preliminary study

G.1.1 Storey response parameters

9S LA 9S LA

9 8 7 6 Storey Floor 5 4 3 2 1

9 8 7

TH
DAP
1st Mode

6 5 4 3 2 1

Uniform

Triangular
Code

0.2 0.4 0.6 Floor Displacement [m]

2 4 6 Interstorey Drift [%]

Evaluation of Pushover Procedures for the Seismic Design of Buildings

220

9S LA

9S LA

9 8 7 6 Storey Storey 5 4 3 2 1

9 8 7

TH
DAP
1st Mode

6 5 4 3 2 1

Uniform

Triangular
Code

2000 4000 6000 8000 Interstorey Shear [kN]

0.5 1 1.5 2 5 Interstorey Moment [kNm] x 10

Fig. 200. 9-Storey frame

20S LA 20 18 16 14 12 Storey Floor 10 8 6 4 2 0 0.2 0.4 0.6 0.8 1 Floor Displacement [m] 20 18 16 14 12 10 8 6 4 2 0

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

2 4 Interstorey Drift [%]

Evaluation of Pushover Procedures for the Seismic Design of Buildings

221

20S LA 20 18 16 14 12 Storey Storey 10 8 6 4 2 0 2000 4000 6000 Interstorey Shear [kN] 8000 20 18 16 14 12 10 8 6 4 2 0

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

1 2 3 5 Interstorey Moment [kNm] x 10

Fig. 201. 20-Storey frame

G.1.2 Bias
9S LA 9S LA

9 8 7 6 5 4 3 2 1 Storey Floor

9 8 7

TH
DAP
1st Mode

6 5 4 3 2 1

Uniform

Triangular
Code

0.5 1 1.5 Floor Displacement

0.5

1 1.5 Interstorey Drift

Evaluation of Pushover Procedures for the Seismic Design of Buildings

222

9S LA

9S LA

9 8 7 6 Storey 5 4 3 2 1 Storey

9 8 7

TH
DAP
1st Mode

6 5 4 3 2 1

Uniform

Triangular
Code

0.5 1 1.5 Interstorey Shear

0.5 1 1.5 Interstorey Moment

Fig. 202. 9-Storey frame

20S LA 20 18 16 14 12 Storey Floor 10 8 6 4 2 0 0.5 1 1.5 Floor Displacement 2 20 18 16 14 12 10 8 6 4 2 0 0.5

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

1 1.5 Interstorey Drift

Evaluation of Pushover Procedures for the Seismic Design of Buildings

223

20S LA 20 18 16 14 12 Storey Storey 10 8 6 4 2 0 0.5 1 1.5 Interstorey Shear 2 20 18 16 14 12 10 8 6 4 2 0

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

0.5 1 1.5 Interstorey Moment

Fig. 203. 20-Storey frame

G.1.3 Dispersion
9S LA 9S LA

9 8 7 6 Storey Floor 5 4 3 2 1

9 8 7

TH
DAP
1st Mode

6 5 4 3 2 1

Uniform

Triangular
Code

0.2

0.4 0.6 0.8 Floor Displacement

0.2

0.4 0.6 Interstorey Drift

0.8

Evaluation of Pushover Procedures for the Seismic Design of Buildings

224

9S LA

9S LA

9 8 7 6 Storey Storey 5 4 3 2 1

9 8 7

TH
DAP
1st Mode

6 5 4 3 2 1

Uniform

Triangular
Code

0.2

0.4 0.6 0.8 Interstorey Shear

0.2

0.4 0.6 0.8 Interstorey Moment

Fig. 204. 9-Storey frame

20S LA 20 18 16 14 12 Storey Floor 10 8 6 4 2 0 0.2 0.4 0.6 0.8 Floor Displacement 1 20 18 16 14 12 10 8 6 4 2 0 0.2

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

0.4 0.6 Interstorey Drift

0.8

Evaluation of Pushover Procedures for the Seismic Design of Buildings

225

20S LA 20 18 16 14 12 Storey Storey 10 8 6 4 2 0 0.2 0.4 0.6 0.8 Interstorey Shear 1 20 18 16 14 12 10 8 6 4 2 0 0.2

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

0.4 0.6 0.8 Interstorey Moment

Fig. 205. 20-Storey frame

G.2

Extensive study Displacement correction factor method

G.2.1 Storey response parameters

9S LA

9 8 7

TH
DAP
1st Mode

6 Storey 5 4 3 2 1

Uniform

Triangular
Code

2 4 6 Interstorey Drift [%]

Evaluation of Pushover Procedures for the Seismic Design of Buildings

226

9S LA

9S LA

9 8 7 6 Storey 5 4 3 2 1 Storey

9 8 7

TH
DAP
1st Mode

6 5 4 3 2 1

Uniform

Triangular
Code

2000 4000 6000 8000 10000 Interstorey Shear [kN]

1 2 Interstorey Moment [kNm]

3 x 10
5

Fig. 206. 9-Storey frame

20S LA 20 18 16 14 12 Storey 10 8 6 4 2 0 2 4 Interstorey Drift [%] 6

TH
DAP
1st Mode
Uniform

Triangular
Code

Evaluation of Pushover Procedures for the Seismic Design of Buildings

227

20S LA 20 18 16 14 12 Storey Storey 10 8 6 4 2 0 2000 4000 6000 Interstorey Shear [kN] 8000 20 18 16 14 12 10 8 6 4 2 0

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

1 2 3 Interstorey Moment [kNm]

4 x 10
5

Fig. 207. 20-Storey frame

G.2.2 Bias
9S LA

9 8 7

TH
DAP
1st Mode

6 Storey 5 4 3 2 1

Uniform

Triangular
Code

0.5

1 1.5 Interstorey Drift

Evaluation of Pushover Procedures for the Seismic Design of Buildings

228

9S LA

9S LA

9 8 7 6 Storey 5 4 3 2 1 Storey

9 8 7

TH
DAP
1st Mode

6 5 4 3 2 1

Uniform

Triangular
Code

0.5 1 1.5 Interstorey Shear

0.5 1 1.5 Interstorey Moment

Fig. 208. 9-Storey frame

20S LA 20 18 16 14 12 Storey 10 8 6 4 2 0 0.5 1 1.5 Interstorey Drift 2

TH
DAP
1st Mode
Uniform

Triangular
Code

Evaluation of Pushover Procedures for the Seismic Design of Buildings

229

20S LA 20 18 16 14 12 Storey Storey 10 8 6 4 2 0 0.5 1 1.5 Interstorey Shear 2 20 18 16 14 12 10 8 6 4 2 0

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

0.5 1 1.5 Interstorey Moment

Fig. 209. 20-Storey frame

G.2.3 Dispersion
9S LA

9 8 7

TH
DAP
1st Mode

6 Storey 5 4 3 2 1

Uniform

Triangular
Code

0.2

0.4 0.6 Interstorey Drift

0.8

Evaluation of Pushover Procedures for the Seismic Design of Buildings

230

9S LA

9S LA

9 8 7 6 Storey 5 4 3 2 1 Storey

9 8 7

TH
DAP
1st Mode

6 5 4 3 2 1

Uniform

Triangular
Code

0.2

0.4 0.6 0.8 Interstorey Shear

0.2

0.4 0.6 0.8 Interstorey Moment

Fig. 210. 9-Storey frame

20S LA 20 18 16 14 12 Storey 10 8 6 4 2 0 0.2 0.4 0.6 Interstorey Drift 0.8 1

TH
DAP
1st Mode
Uniform

Triangular
Code

Evaluation of Pushover Procedures for the Seismic Design of Buildings

231

20S LA 20 18 16 14 12 Storey Storey 10 8 6 4 2 0 0.2 0.4 0.6 0.8 Interstorey Shear 1 20 18 16 14 12 10 8 6 4 2 0 0.2

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

0.4 0.6 0.8 Interstorey Moment

Fig. 211. 20-Storey frame

G.3

Extensive study Scaling factor method

G.3.1 Storey response parameters

9S LA

9 8 7

TH
DAP
1st Mode

6 Storey 5 4 3 2 1

Uniform

Triangular
Code

2 4 6 Interstorey Drift [%]

Evaluation of Pushover Procedures for the Seismic Design of Buildings

232

9S LA

9S LA

9 8 7 6 Storey

9 8 7 6 Storey 5 4 3 2 1

TH
DAP
1st Mode
Uniform

5 4 3 2 1

Triangular
Code

5000 10000 Interstorey Shear [kN]

0.5 1 1.5 2 Interstorey Moment [kNm]

2.5 x 10
5

Fig. 212. 9-Storey frame

20S LA 20 18 16 14 12 Storey 10 8 6 4 2 0 2 4 Interstorey Drift [%] 6

TH
DAP
1st Mode
Uniform

Triangular
Code

Evaluation of Pushover Procedures for the Seismic Design of Buildings

233

20S LA 20 18 16 14 12 Storey Storey 10 8 6 4 2 0 2000 4000 6000 8000 Interstorey Shear [kN] 20 18 16 14 12 10 8 6 4 2 0

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

1 2 3 4 5 Interstorey Moment [kNm] x 10

Fig. 213. 20-Storey frame

G.3.2 Bias
9S LA

9 8 7

TH
DAP
1st Mode

6 Storey 5 4 3 2 1

Uniform

Triangular
Code

0.5

1 1.5 Interstorey Drift

Evaluation of Pushover Procedures for the Seismic Design of Buildings

234

9S LA

9S LA

9 8 7 6 Storey 5 4 3 2 1 Storey

9 8 7

TH
DAP
1st Mode

6 5 4 3 2 1

Uniform

Triangular
Code

0.5 1 1.5 Interstorey Shear

0.5 1 1.5 Interstorey Moment

Fig. 214. 9-Storey frame

20S LA 20 18 16 14 12 Storey 10 8 6 4 2 0 0.5 1 1.5 Interstorey Drift 2

TH
DAP
1st Mode
Uniform

Triangular
Code

Evaluation of Pushover Procedures for the Seismic Design of Buildings

235

20S LA 20 18 16 14 12 Storey Storey 10 8 6 4 2 0 0.5 1 1.5 Interstorey Shear 2 20 18 16 14 12 10 8 6 4 2 0

20S LA

TH
DAP
1st Mode
Uniform

Triangular
Code

0.5 1 1.5 Interstorey Moment

Fig. 215. 20-Storey frame