DAPS2015 Abstracts

Table of Contents
Welcome Message 1
Local Organising Committee 3
International Committees 4
Conference Information 5
Programme at a Glance 6
th
Day 1 Programme (19 May 2015) 7
Day 2 Programme (20th May 2015) 10
Day 3 Programme (21st May 2015) 13
Keynote Papers
- Recent Observations on Design and Analysis of Protective Structures 18
Prof Theodor Krauthammer
- Numerical Studies on Reinforced Concrete Sub-Structures under 29
Progressive Collapse Caused by Contact Detonation
Prof Tan Kang Hai
- Analysis and Design for Conventional Structures under Extreme Multi- 46
hazard Loading Conditions
Prof Phillip L. Gould
- Understanding the Boundary Condition when Designing Protective 57
Structures: The Case of Fires and Explosions
Prof José L. Torero
- Structures Subjected to Extreme Loading Conditions. Do We Need New 67
Knowledge?
Prof Magnus Langseth
- Design of Aircraft Structures for Protection of Bird Strike 69
Prof Yulong LI
Oral Presentations
- Blast, Shock and Impact Load Abstracts 74
- Impact and Explosion Tests Abstracts 92
- Material Behaviour under Blast 104
- Numerical Modelling and Simulation Abstracts 112
- Protection Concepts and Design Abstracts 140
- Rock Dynamics Abstracts 158
- Structural Behaviour and Damage 166
- Structures under Fire Abstracts 186
Poster Presentations Abstracts 194
Authors’ Index 210
Exhibition Floorplan 215
Exhibitors’ List 216
WELCOME MESSAGE
On behalf of the Local Organizing Committee, we are delighted to welcome all the delegates and their
guests to Singapore, for the Fifth International conference on Design and Analysis of Protective
Structures (DAPS2015) on 19-21 May 2015.
This conference proceeding contains a collection of 120 technical papers, which were selected by the
International Scientific Committee from 200 abstracts from 24 countries. After the review and
selection process, a range of papers from 20 countries were selected and will be delivered at the
conference through 6 keynotes and 99 papers, in presentations, and 15 additional papers displayed in
posters.
The first DAPS conference was originally called Design and Analysis of Protective Structures against
Impact/Impulsive/Shock Loads (DAPSIL), and was held in Japan in December 2003; it was
conceived by three founding institutes: the National Defense Academy (Japan), the Pennsylvania
State University (USA), and the Nanyang Technological University (Singapore). At the time, there
were three founding co-chairmen, Prof. Tomonori Ohno, Prof. Theodor Krauthammer, and Prof. Pan
Tso-Chien. Since then DAPS has been held in Singapore twice, followed by South Korea before
returning to Singapore for this 5th edition.
The DAPS series has broadened its coverage over the years. DAPS2015 covers very diversified
research areas, including: Protection Concepts and Design; Blast, Shock and Impact Load; Structural
Behaviour and Damage; Material Behaviour under Blast; Impact and Explosion Tests; Numerical
Modelling and Simulation; and Structures Under Fire. Nonetheless all these research areas have the
common objective of providing protective robust and resilient Building Infrastructure solutions
against the threats and accidents of this age.
We hope that DAPS2015 will provide a platform for the exchange and sharing of knowledge on the
latest developments in protective technology. DAPS is rooted in the Asia Pacific region but welcomes
contributions and participants from all over the world. We hope the new knowledge shared in this
conference will benefit researchers and protective engineers, as well as policy makers.
DAPS2015 is jointly organised by Nanyang Technological University (NTU) and the Defence
Science & Technology Agency (DSTA), Singapore. We would like to thank the local organizing
committee (CHAI Gin Boay, KANG Kok Wei, LEE Chi King, LIM Heng Soon, Edwin LIM Tze
Meng, Debbie LOW, LU Guoxing, ZHOU Hongyuan ) for their hard work, the International Advisory
Committee for their guidance, the International Scientific Committee for their help in paper reviews,
1
WELCOME MESSAGE
and CMA International Consultants for their excellent logistics support. We also like to thank all the
conference sponsors for their support.
We are sincerely grateful to all the presenters for sharing their research findings at DAPS2015, and
we wish all participants a fruitful conference and a pleasant stay in Singapore.
ZHAO Zhiye Leonard HENG Eu Chang

Chairman Co-Chairman
DAPS2015 Local Organising Committee DAPS2015 Local Organising Committee
2
LOCAL ORGANISING COMMITTEE
Chairman:
ZHAO Zhiye, Nanyang Technological University, Singapore
Co-Chairman:
Leonard HENG Eu Chang, Defence Science & Technology Agency, Singapore
Committee Members:
CHAI Gin Boay, Nanyang Technological University, Singapore
KANG Kok Wei, Defence Science & Technology Agency, Singapore
LEE Chi King, Nanyang Technological University, Singapore
LIM Heng Soon, Defence Science & Technology Agency, Singapore
Debbie LOW, Nanyang Technological University, Singapore
LU Guoxing, Nanyang Technological University, Singapore
ZHOU Hongyuan, Nanyang Technological University, Singapore
3
INTERNATIONAL COMMITTEES
International Advisory Committee

1. Theodor KRAUTHAMMER, University of Florida, United States
2. Masuhiro BEPPU, National Defense Academy, Japan
3. Magnus LANGSETH, Norwegian University of Science and Technology, Norway
4. LIM Chee Hiong, Nanyang Technological University, Singapore
5. ONG Yew Hing, Defense Science and Technology, Singapore
6. Hong-Yul PAIK, Agency for Defense Development, Korea
7. PAN Tso-Chien, Nanyang Technological University, Singapore
8. TAN Kang Hai, Nanyang Technological University, Singapore
9. Klaus THOMA, Ernst Mach Institute, Germany
International Scientific Committee

1. Daniel AMBROSINI, National University of Cuyo, Argentina
2. Nemkumar BANTHIA, The University of British Columbia, Canada
3. Gabi BEN-DOR, Ben-Gurion University of the Negev, Israel
4. Kenneth CARPER, Washington State University, United States
5. Karen Mary CHONG Oi Yin, Defense Science and Technology Agency, Singapore
6. Nawawi CHOUW, The University of Auckland, New Zealand
7. John CRAWFORD, Karagozian & Case, United States
8. Xiuli DU, Beijing University of Technology, China
9. Norbert GEBBEKEN, University of the German Armed Forces, Germany
10. Hong HAO, Curtin University, Australia
11. Theodor KRAUTHAMMER, University of Florida, United States
12. Magnus LANGSETH, Norwegian University of Science and Technology, Norway
13. Qingming LI, University of Manchester, United Kingdom
14. Zhong-Xian LI, Tianjin Chengjian University and Tianjin University, China
15. Priyan MENDIS, The University of Melbourne, Australia
16. Gerald NURICK, University of Cape Town South Africa
17. Jong-Yil PARK, Seoul National University of Science and Technology, Korea
18. SEAH Chong Chiang, Defense Science and Technology Agency, Singapore
19. Yoshimi SONODA, Kyushu University, Japan
20. Klaus THOMA, Ernst Mach Institute, Germany
21. David YANKELEVSKY, Technion-Israel Institute of Technology, Israel
22. Qingming ZHANG, Beijing Insitute of Technology, China
4
CONFERENCE INFORMATION
Conference Venue
Furama Riverfront Hotel
405 Havelock Road
Singapore 169633
Official Language
The Official Language of the Conference is English.
Conference Materials
Upon registration, you will receive your Conference materials and a namebadge, which
should be worn at all times during the Conference functions.
Conference Delegates and Entitlements

All registered Conference delegates are entitled to:
 Attendance to all Conference sessions and Exhibition
 Conference proceedings (in thumbdrive)
 Daily morning and afternoon teabreaks
 Daily lunch
 Conference Banquet
Liability
The Conference Organisers are not liable for personal accidents or loss/damage of private
properties of registered delegates during the Conference. Registered delegates are encouraged
to make their own arrangements for personal insurance.
Disclaimer
Whilst every attempt will be made to ensure that all aspects of the Conference mentioned will
be as scheduled, the Conference Organisers reserves the right to make changes should the
need arises. The Conference Organisers is not responsible for any personal expenses incurred
or any loss suffered by any delegate or his guest in connection with the Symposium.
Conference Banquet
In conjunction with the Conference, a Conference banquet will be held on Wednesday, 20th
May 2015 at
Chui Huay Lim Club

Level 2 Grand Ballroom
190 Keng Lee Road
Singapore
The dress code for the dinner is business casual (tie not required). Please bring along your
Conference namebadge for admission.
There will be a 2-way bus transfer to transport delegates to the dinner venue from Furama
Riverfront Hotel at 6.30pm.
5
PROGRAMME AT A GLANCE
6
DAY 1 PROGRAMME (19th MAY 2015)
Time Tuesday, 19th May 2015
0800 - 0830 Registration @ Venus Ballroom Foyer
0830 - 0900 Opening Ceremony @ Venus Ballroom I & II
Keynote – Recent Observations on Design and Analysis of Protective Structures

Theodor KRAUTHAMMER
0900 - 0940
Session Chair: Zhiye ZHAO
Venus Ballroom I & II
Keynote – Numerical Studies on Reinforced Concrete Sub-Structures under

Progressive Collapse Caused by Contact Detonation
0940 - 1020 TAN Kang Hai
Session Chair: Zhiye ZHAO
Venue: Venus Ballroom I & II
1020 - 1050 Teabreak & Poster Viewing @ Jupiter Ballroom
Material Behaviour under Blast Numerical Modelling and Simulation I

Session Chair: John CRAWFORD Session Chair: Sritawat KITIPORNCHAI
Venus Ballroom III Venus Ballroom I & II
A Micromechanical Method to Predict

Numerical Study on Ballistic Impact of
Macroscopic Behaviour of Progressive
1050 - 1110 Perforated Armour Plates
Failure in Brittle Rock
Qian SHI
Xiaozhao LI
Reaction Force and Delamination Design and Optimization of Laminated

Properties of Laminated Glass Windows Transparent Armor by Finite Element
1110 - 1130
subject to Blast Loads Analysis
Paolo DEL LINZ Yukolthorn SRILOY
Finite Element Analysis of Dynamic

Damping Characteristics of Sand-Rubber
Crushing of Functionally Graded Foam-
1130 - 1150 Chip Mixtures as a Seismic Barrier
filled Square Columns
Zhuoyuan CHENG
Xuehui YU
Strain Rate Effects of Compressive Structural Optimization of K4-rating

Properties of Hollow Glass Microspheres Removable Anti-ram Bollards Based on
1150 - 1210
filled Polymers Orthogonal Design Method
Xin ZHANG Suwen CHEN
Prediction of Penetration Depth of

Design and Analysis of Protective
Projectiles into Concrete Based on
1210 - 1230 Structures: A Fresh Look at Concrete
Support Vector Machines
Peter WEBER
Qiang PAN
Lunch @ Mercury Ballroom

1230 -1330
7


Blast, Shock and Impact
Protection Concepts and Numerical Modelling and
Load I
Design I Simulation II
Session Chair: Young-Soo
Session Chair: Youjun NING Session Chair: Gin Boay CHAI
YOON
Venus Ballroom III Venus Ballroom I
Venus Ballroom II
Performance Based Numerical and Finite Element Analysis of

Framework for Resilient Experimental Investigation Long Cylindrical Tubes
1330 - 1350 Design of Structures to on Hail Impact on Subjected to Internal
Blast Loads Composite Panels Explosions
Serdar ASTARLIOGLU Tao SUO Guoxing LU
Low-velocity Impact
Blast Mitigation by
Response of Geometrically
Guidelines for Level of Stiffened Composite Plates
Asymmetric Sandwich
1350 -1410 Protection of Facility for Vehicles under Mine
Beams with Local Denting
Jong Yil PARK Blast
Effect
M.D. GOEL
Jianxun ZHANG
An Analytical Model for

Finite Element Analysis of
Risk Protec CI- Research Low-Velocity Impact
FRP-strengthened
Project for Aircraft Impact Response of Metal
1410 - 1430 Reinforced Concrete
Funded by EU Sandwich Beams with a
Panels under Blast Loading
Alexander SIEFERT Foam Core
Xiaoshan LIN
Chunping XIANG
Quantitative Risk Analysis

of Gas Explosions in Development of a New Effect of Blast Load on
Tunnels; Probability, Class of Penetration Response of Concrete
1430 -1450
Effects and Consequences Analysis Methods Rectangular Storage Tanks
Martijn VAN DER John CRAWFORD Bojan ZLENDER
VOORT
An Engineering Approach
Environmentally Low Fast Analysis of Safety
for Risk, Resilience and
Impact Tunnel Blasting in Areas for EOD in Urban
1450 -1510 Vulnerability Assessment
Urban Residential Area Environment
of Urban Areas
Keita IWANO Stefan GREULICH
Kai FISCHER
8


Blast, Shock and Impact
Protection Concepts and Structural Behaviour and
Load II
Design II Damage I
Session Chair: Masuhiro
Session Chair: Jose TORERO Session Chair: Xinzheng LU
BEPPU
Venus Ballroom III Venue: Venus Ballroom I
Venus Ballroom II
Joint Slippage Effects on
Local Damage of Fiber
Analysis of Explosion the Impact Dynamic
Reinforced Cementitious
Incidents in Mass- Response of Lattice Angle
1540 - 1600 Composite Plates subjected
Transport Systems Steel Tower Subject to
to High Velocity Impact
Gero MEYER Break of Conductor
Masuhiro BEPPU
Wenqiang JIANG
A Structured Approach to Local Effects due to the
Forensic Study of Opening Sizes on the
Impact Resistance of Sand
Explosions: The TNO Negative Bending Regions
against Projectile
1600 - 1620 Inverse Explosion Analysis of Steel-Concrete
Penetration
Tool Composite Beam owing to
Siau Chen CHIAN
Martijn VAN DER Axial Compression
VOORT Mahesan BAVAN
Single Impact and
Residual Strength Tests on An Efficient Analytical
Buildings as Barricades Two-way Concrete Panels: Failure Approach in the
1620 - 1640 Model Effects of Adding Steel Global Analysis of a
Michael SWISDAK Fiber and Changing Rebar Buried Pipeline System
Detail Kwong Ming TSE
Ki Yeon KWON
Energy Absorption
Reducing Land Usage for
Performance of Ni-based
Blast Protected JTC
Superalloy Honeycomb
1640 - 1700 Standardised 66/22kV
Structure under Different
Substations
Ball Velocities
Peter J HOAD
Weiguo GUO
Optimization of Material
A Novel Device for Blast
Systems with
1700 - 1720 Energy Dissipation
Magnetorheological Fluids
Hongyuan ZHOU
Adam WISNIEWSKI
9


Time Wednesday, 20th May 2015
Keynote - Analysis and Design for Conventional Structures under Extreme Multi-
hazard Loading Conditions
0900 - 0940 Phillip L. GOULD
Session Chair: Theodor KRAUTHAMMER
Keynote - Understanding the Boundary Condition when Designing Protective
0940 - 1020 José L. TORERO
Session Chair: Theodor KRAUTHAMMER
Venus Ballroom I and II
Structural Behaviour and Damage II Blast, Shock and Impact Load III
Session Chair: Chi King LEE Session Chair: Hongyuan ZHOU
Venus Ballroom I & II Venus Ballroom III
Structural Behaviour of Precast Beam-

Numerical Predictions of the Negative
Column Sub-Assemblages Column
1050 - 1110 Phase
Removal Scenarios
Piotr SIELICKI
Kang Hai TAN
Progressive Collapse Performance of
Blast Load Assessment for Safety in Urban
Three-Dimensional Concrete Frames
1110 – 1130 and Enclosed Environments
Subjected to Corner Column Removal
David STEVENS
Jun YU
Feasibility of FRP Composite Material
Blast Effects inside a Partially Enclosed
for Building Protection against
1130 – 1150 Carpark
Progressive Collapse
Choon Keat ANG
Kai QIAN
Assessment of Progressive Collapse
Coupled CFD/CSD Simulations of Dust
induced by Blast using Updated Single
1150 -1210 Production by Fragmenting Charges
Column Removal Scenarios
Joseph BAUM
Anh Tuan PHAM
Local Damage of RC Beams according
Approach to Combined Blast and
to Variation of Shear Reinforcement
1210 - 1230 Fragmentation Modelling
after Blast Loading
Joel Ke Wei ONG
Jinyoung LEE

1230 -1330
10



Impact and Explosion
Structural Behaviour and Numerical Modelling and
Tests I
Damage III Simulation III
Session Chair: Jaap
Session Chair: Zilong ZHOU Session Chair: Phillip GOULD
WEERHEIJM
Venus Ballroom I Venus Ballroom II
Venus Ballroom III
Experimental and
Theoretical Analysis of the Verification and Validation Fibre Reinforcing
Progressive Collapse of Material Models for Composite Piping Systems
1330 - 1350 Resistance of One-way Design of Protective against Impact
Reinforced Concrete Structures Dexin XIONG
Beam-Slab Substructures David MORIN
Xinzheng LU
Extended Beam Element

Formulation based on On the Effectiveness of Small Scale Vapour Cloud
1350 -1410 Static Condensation to Blast Mitigation with Soft Explosion Tests
Simulate Rotational Resistance Claddings Yadong LIN
Discontinuity Hongyuan ZHOU
Xu LONG
Bond Characteristics of
Blast Performance of High Explosive Trial to
Concrete-to-SHCC
Strength Concrete Panels: Investigate the Hazard
Interfaces and Numerical
Experimental, Numerical Mitigation Effects of Earth
1410 - 1430 Analysis of SHCC-
and Analytical Cover
strengthened RC Slabs
Investigation Kok Wei KANG
under Blast Loading
Satadru DAS ADHIKARY
Rainnevo Chandra LADO
The Behaviour of Steel and Observations from Blast-

Blast Performance of RC
Composite Beam-Column Induced Liquefaction Tests
1430 -1450 Slabs Subjected to
Joints under a middle- in Singapore
Adjacent and Contact Blast
Column Removal Scenario Sarma ANAND
Marek FOGLAR
Bo YANG
Investigation of Dynamic
Deformation Behaviours in
Effect of Torsion on 2D Damage Assessment of
Structural Steels at
RC Frame under Middle Reinforced Concrete
1450 -1510 Intermediate Strain Rates
Column Removal in Frames under Progressive
using Newly Constructed
Progressive Collapse Collapse
Impact Tensile Test
Chi King LEE Jian WENG
Aparatus
Hyung-Seop SHIN
11



Protection Concepts and Impact and Explosion
Numerical Modelling and
Design III Tests II
Simulation IV
Session Chair: Magnus Session Chair: Chong Chiang
Session Chair: Guoxing LU
LANGSETH SEAH
Venus Ballroom II
Venus Ballroom III Venus Ballroom I
Design, Analysis and Numerical Modelling of Experimental Study of

Testing of Blast-Resistant Projectile Explosive Non-composite SCS
1540 - 1600 Charge Damage during Sandwich Panels under
Curtain Wall
Shengrui LAN Penetration Impulsive Loading
Youjun NING Yonghui WANG
The Use of Fiber

Composites for Blast and Blast Tests on Sandwich
Fragmentation Protection Energy Based Load-
1600 - 1620 Beams with Thin-walled
of Selected Civilian Impulse Diagrams
Tube Cores
Infrastructure Ying-Kuan TSAI
Xinmei XIANG
Moshe RAVID
Nested Tube System Effect of Seismic Barrier Experimental Investigation

Applicable to Protective on Response of of Response of Steel Plates
1620 - 1640 Structures against Blast Underground Structures due to Close-in Blast
Shock Subjected to Ground Loading from Spherical
Ze Liang YU Shock Liquid Explosive
Jia Han CHEW Alex REMENNIKOV
Improving Robustness of
Medium-Rise Composite Performance of Pile Lesson from Blast Test of
1640 - 1700 Building using Multiple Groups Subjected to Steel Plate in Elastic
Belt-Truss Systems Surface Explosion Domain
Chan Ghee KOH Zhiye ZHAO Jong Yil PARK
Photonic Doppler
Velocimetry
Performance of an Instrumentation in Real
Explosive Water-mist Dynamic Response of
Scale Concrete Slab Blast
Spray System for Dust Water Landing Capsule
1700 - 1720 Loading Experiments with
Control in Explosive Body based on the Mesh-
an Attempt to Measure the
Demolition free Method
Influence of the Type of
Hyon Soo KIM Yonghu WANG
Concrete on the Debris
Velocity
Martin KUNZEL
1900 Conference Banquet
12
DAY 3 PROGRAMME (21st MAY 2015)
Time Thursday, 21st May 2015
Keynote - Structures Subjected to Extreme Loading Conditions. Do We Need New

Knowledge?
0900 - 0940 Magnus LANGSETH
Session Chair: Leonard HENG
Keynote - Design of Aircraft Structures for Protection of Bird Strike
Yulong LI
0940 - 1020
Session Chair: Leonard HENG
Structural Behaviour and Damage IV Numerical Modelling and Simulation V

Session Chair: Karen CHONG Oi Yin Session Chair: Wancheng ZHU
Modelling Dynamic Tensile Failure of

Dynamic Behaviour of UHPC Mixed
1050 - 1110 Quasi-brittle Materials using Stress-
with Nano Material
enhanced Nonlocal Models
Jun LI
L.F. PEREIRA
Experimental Frequency Domain

Development of Modular Passive
1110 - 1130 Assessment of Direct Shear in NSC and
Attenuation Systems
UHPC
Hui Wei YEO
Theodor KRAUTHAMMER
Response to RC Slabs subjected to Assessment of Fast-Running Algorithms

1130 - 1150 Combined Blast and Fragment Loading for Tunnel Blast Predictions
Rickard FORSEN Teng Sheng PEH
End Restraint Effect on the Dynamic

Numerical Modelling of Chemical
1150 - 1210 Response and Failure Behaviour of Steel
Dispersion via Elevated Air Intake
Beams Subjected to Pulse Loading
Wee Teik KHOO
Feng XI
An Assessment Protocol for Reinforced
Concrete Columns Subject to Internal A Cellular Automaton Method for 2-
1210 - 1230 Building Detonations dimansional Solid Mechanics
Gayan Lakshitha Wijesundara Weifeng YUAN
MUDALIGE

1230 -1330
13
DAY 3 PROGRAMME (21st MAY 2015)
Time Thursday, 21st May 2015

Structures under Fire Rock Dynamics
Session Chair: Yulong LI Session Chair: Jianchun LI
Application of Cellular Automaton in Dynamic Properties of Water Saturated

1330 - 1350 Fire Engineering Sandstone
Weifeng YUAN Zilong ZHOU
Using Performance-based Fire Numerical Study of Dynamic Fluid Flow in

1350 -1410 Engineering Design in a Protected Rock Fracture Network using the Lattice
Facility Boltzmann Method
Yong Kiang TAN Peijie YIN
Analytical Evaluation of Bi-

Directionally Prestressed Concrete Panel Dynamic Ring Test of Rock under
1410 - 1430 Behaviour under Blast-Fire Combined Diametral Impact Load by SHPB
Loading using Parametric Study Diyuan LI
Seung Jai CHOI
Effect of Braces on the Progressive Numerical Simulation on Failure Process
1430 -1450 Collapse Mechanism of Steel Structures of Unloaded Rock Triggered by Dynamic
Exposed to Fire Disturbance
Jian JIANG Wancheng ZHU
Design of Wireless System for the

Blast Wave Propagation across Jointed
Detection of Accidental and Terrorist
1450 -1510 Rock Masses and its Influence on the
Explosions and Fires in Manufacturing
Ground Motion
Sites
Jianchun LI
Shalva MARJANISHVILI
1510 - 1530 Tall Buildings in Post Earth Quake Fire

Tadepalli MADHAVI
1530 – 1600 Teabreak & Poster Viewing @ Jupiter Ballroom
1600 - 1630 Closing Ceremony & Best Paper Award Presentation

14
KEYNOTE PAPERS
Tuesday, 19th May 2015

0900 - 0940 hours
Recent Observations on Design and Analysis of Protective Structures

Theodore Krauthammer
Center for Infrastructure Protection and Physical Security (CIPPS), University of Florida, USA
(tedk@ufl.edu)
Recent studies have focused on both active and passive protection, the definition and
characterization of combined blast and fragment loads on structural elements, and the influence of
Ultra High Performance Concrete (UHPC) on the behavior of structural elements. These studies have
highlighted important issues that have not been addressed in various design guidelines. This paper is
aimed at describing several such issues, and proposes approaches that could be used to address them.
Keywords: Impact; Blast; Shock; Fragments; Impulsive; Load; Concrete materials; Active; Passive;
Protective analysis and design; Testing; Numerical simulations
1. Introduction
Protected facilities are expected to perform their function under very severe dynamic loading
environments, as extensively described in the literature [1-8]. These loading environments have
continuously become more severe, as directly related to the evolution in weapon systems and the
escalation in terrorism capabilities. Consequently, one is required to enhance the levels of protection
by employing increasingly more sophisticated and more expensive approaches. From a structural
engineering standpoint, one needs to satisfy the following underlying equation of equilibrium
between the applied loads and a structure’s ability to resist them:
(1)
This equation could represent either a single-degree-of-freedom (SDOF) approach or an

advanced analysis approach (e.g., finite element or multi-Physics programs). In which, M is the mass
(equivalent or mass matrix), C is damping (coefficient or matrix), K is stiffness (coefficient or
matrix), and F is the load (single parameter or vector). , , are the acceleration, velocity, and
displacement (at a point or vectors). One should note that Eq. (1) needs to be adjusted to
accommodate combinations of material and geometric nonlinearities, and represent time-, and/or
rate-dependent changes in any of these parameters. In SDOF approaches, the term Ku can be replaced
by the nonlinear resistance R that must be derived for the structural system under consideration.
Traditionally, structural/protective engineers have attempted to adopt approaches that could
affect both sides of the equilibrium equation by reducing the magnitude of the applied loads, and/or
enhancing the structural resistance. Typically, such activities were passive; e.g., increase the standoff
distance to the explosive device, or shield the structure, to reduce the load magnitude, and/or use
stronger structural elements (with combinations of innovative materials, dimensions, and detailing)
to enhance the resistance. In recent years, however, also active means have been introduced to further
reduce the magnitude of the load (either by intercepting the ordnance, or by suppressing the blast
effects).
Several observations have been selected from recent and/or ongoing research activities to
address the loading and structural resistance issues, and they are presented in separate sections,
below. Finally, these observations are discussed to explore possible implications on the evolution of
protecting critical infrastructure systems in the future.
18
KEYNOTE PAPERS
2. Observations Related to Explosively-Induced Loads
2.1 Combined blast and fragment effects
Current design manuals and related practical explosive load simulation tools, as mentioned
earlier, consider only the blast for loading the structure, while fragments are considered only for
penetration. However, experimental and numerical studies have been conducted to understand the
synergistic effects of a combined blast and fragment loading [9-11]. Although, due to the complicated
Physical phenomena, this synergetic combination is still not well understood, test data have shown
that cased charges with various case weight combinations that had the same amount of explosive
material delivered to targets at different standoff distances up to four times the amount of impulse, as
compared with the same bare charge at the same standoff distances. A recent study attempted to use
the same manuals and tools, but to extract a loading component from the penetrating fragments with
the appropriate times of arrival and durations [12]. An example from that study can be used to both
illustrate this phenomenon and the effectiveness of the adopted approach, as shown in Fig. 1. The
impulse due to pure blast from a 0.98 kg CompB charge recorded in the test (MK0) is plotted as the
orange solid line. The impulse due to synergetic effects of both fragmentation and blast recorded in
test (MK4) for a 0.98 kg CompB charge and a 4 kg case is plotted as the green solid line. The impulse
due to pure blast and the synergetic effects obtained in the proposed procedure are plotted as blue
dash line and red dash line, respectively. It should be noted that the impulse caused by pure blast
effect in the proposed procedure is obtained by taking an average of the spherical and hemispherical
burst. Clearly, the combined blast-fragment loading effects are much more severe because they
include the pressure on the target from both the blast and fragments, and one must address also the
added damage due to fragment penetration. Therefore, current design guidelines should be updated to
enable analysts and designers to address this much more severe loading combination.
Fig. 1: Measured and simulated impulse vs. range for cased and uncased charges [12]
2.2 Blast load suppression
2.2.1 Thermodynamic suppression

Blast suppression by shielding the target with a water layer (either in liquid form or water mist)
has been of interest for many years. This concept was been studied by several research teams (e.g.,
[13-18]), and the results showed some promise. The available data indicated that using a water mist
barrier between the blast source and the target could reduce the peak pressure by about 30% - 58% (a
62% attenuation was achieved with a foam suspension mist [16, 19, 20]), as shown in Fig. 2. A
careful analysis of those studies showed that the main source of pressure reduction was based on
19
KEYNOTE PAPERS
momentum transfer between the shock wave and the water droplets, and the mechanical energy used
to break them up into smaller droplets. Although very encouraging, the approach needs to be
developed further to achieve a full Thermodynamic suppression of the blast that might be able to
approach an overpressure reduction factor of between 3.5 and 9.8 that was achieved in small scale
laboratory tests [21,22].
(a) Test set up
(b) overpressure histories, 1 - without mist, 2 - with water mist

Fig. 2: Blast suppression test with, W = 2 kg charge [20]
2.2.1 Mechanical load transfer limiters

Mechanical load transfer limiters have been used extensively for protecting critical
infrastructure facilities against seismically-induced ground motions. Similar approaches were
adopted also for the protection from ground shock produced by nuclear detonations. A recent study
[23] has shown that placing tailored shock absorbers between the building envelop and the structural
framing system could reduce the forces transmitted to the frame by about 50%. Additional reductions
of the load transfer could be achieved by designing the building envelop as a structural system whose
support reactions are modulated by the envelope’s dynamic behavior [24].
3. Observations on Enhancing Structural Resistance
3.1 Potential applications of Ultra High Strength Concrete (UHPC)
As the protection requirements evolve to address the enhanced loads associated with more
severe threats, engineers have explored the use of innovative materials for protective design. One
such option is adopting ultra high strength concrete (UHPC) for structural concrete protected
facilities. Ultra-High-Performance Concrete (UHPC) is a relatively new cementitious material which
20
KEYNOTE PAPERS
uses a highly optimized cement matrix consisting of very fine aggregates and fibers to achieve
extremely high compressive and tensile strengths, as compared to the properties of normal-strength
concrete (NSC) and high-strength concrete (HSC). In addition to higher strength, UHPC offers
higher ductility and durability as compared to NSC or HSC. However, several issues may hinder a
wide-spread adoption of UHPC. Commercially available UHPCs are proprietary mixes, and the
material cost alone is more than ten-fold the cost of NSC. Then, UHPC requires stringent preparation
and curing practices, which add considerably to the cost and often prohibit its use in cast-in-place
type applications. Finally, there is a big gap in the state-of-knowledge on how to analyze and design
structures using this innovative material, and how such structures will respond to static, dynamic, or
impulsive loads. For this purpose, significant research needs to be done in order to extend the
knowledge obtained, from component level (e.g., cylinder, beam, panel, etc.) testing to guidelines
and principles for structural analysis and design. To achieve these goals, one must develop a family of
validated and reliable material models for high-fidelity finite element and fast-running analysis
methods, as well as well-defined design guidelines. An ongoing multi-year research effort is aimed at
addressing various aspects of this general problem [25-33], and several observations from it are
briefly presented next.
3.1.1 Material behavior of cylinders

A series of about 38 MPa NSC and two types of up to 200 MPa UHPC (UHPFRC with and
UHPC w/o steel fibers) cylinders were tested under static and impact loads, and the data were
extensively studied. All NSC and UHPC materials were developed at the US Army Engineer
Research and Development Center (ERDC), and were provided for these studies. Follow up
numerical simulations were used to define the behavioral differences related to the specimens’
material types, geometries, and loading rates. The corresponding findings were used to extract
parameters for describing the observed behaviors with appropriate material models. These material
models were then used also to simulate the structural behaviors of various beams, columns, slabs, and
direct shear specimens to further refine the material models, as discussed in several other sections of
this article. The structural behavior simulations were done with both fast-running approaches and
advanced finite element codes. An example of data from one of the tests and simulations on a ϕ100
mm X 200 mm UHPFRC specimen with steel fibers is shown in Fig. 3.
(a) Observed failure (b) Finite element model
(c) Measured data (d) Simulation vs. test data

Fig. 3: Test and simulation data for a ϕ100 mm X 200 mm UHPFRC specimen with steel fibers
21
KEYNOTE PAPERS
These activities have shed new light on the behaviors of NSC, UHPC (UHPFRC with and
UHPC w/o steel fibers) cylinders under different combinations of materials, geometries and loading
rates. That has enabled a better understanding of the Physical phenomena that underline these
behaviors, and the derivation of more accurate material models for both fast-running and advanced
numerical simulations. Currently, this effort is still going on to address the coupled size- and
rate-effects in these specimens, and more information on it will be presented in an accompanying
paper.
3.1.2 Flexural behavior of beams

A series of NSC and two types of UHPC (UHPFRC with and UHPC w/o steel fibers) beams
(using the same mixes, as for the cylinders) with different reinforcement details were tested under
static and impact loads, as shown in Fig. 4 and Table 1, and the data were extensively studied.
Fig. 4: Test beam details (units: mm)
Table 1: Beam designations and properties

Reinforcement ratio, %
Designation Material
Tension Compression Shear
B1 NSC 2.00 1.47 0.73
B2 NSC 2.00 1.47 0.00
B3 UHPFRC 2.00 1.47 0.73
B4 UHPC 2.00 1.47 0.73
B5 UHPFRC 0.00 0.00 0.00
B6 UHPC 2.00 1.47 0.00
Addressing the dynamic behavior of these beams can highlight the potential of using UHPC
materials for protective design. For all tests, the drop mass was 408.6 kg, and the drop height was
between 3.91 m and 3.96 m. Typical data obtained from these tests included the impact force vs. time,
mid-span and quarter-span displacements vs. time, reaction forces vs. time, and steel strains vs. time
at several locations along the tensile and compressive reinforcement bars, as illustrated in Fig. 5. A
summary of representative data is provided in Table 2.
22
KEYNOTE PAPERS
2000
1750
1500
1250
Load, kN 1000
750
500
250
0
30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
-250
-500
Time, ms
0.16
0.08
0
Strain, %
30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
-0.08
-0.16
-0.24
-0.32
-0.4
Time, ms
100
Midspan displacement, mm
90
80
70
60
50
40
30
20
10
0
30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Time, ms
Fig. 5: Typical beam impact test data
Table 2: Typical test data from impact beam tests
Test Peak Load Peak Midspan Disp Peak Reaction

kN mm. mm
B1 1311.3 65.6 225.5

B2 1458.6 179.8 150.8
B3 1770.4 67.7 369.6
B4 1226.8 79.8 328.3
B5 1841.1 190.5 261.1
B6 1834.9 54.4 447.5
Follow up numerical simulations with both fast-running and detailed advanced finite element
approaches that included the material models derived from the study on the cylinders were used for
comparisons with test data, and to define the parameters that affected the observed behaviors. In the
static domain, the best correlation between the experimental and numerical results was observed for the
NSC beams that had both flexural and shear reinforcement. The correlation between the results in the NSC
beams that had only flexural reinforcement and no shear reinforcement were very poor, primarily due to
23
KEYNOTE PAPERS
the brittle and non-symmetric nature of the failure. The numerical and experimental results for the UHPC
beams were generally in good agreement, with the exception of unreinforced UHFRPC beams. In the
dynamic domain, the differences between the test and numerical simulations were between 3% and 13%
for the NSC beams, between 2% and 11% for the UHPC beams with both longitudinal and shear
reinforcement and steel fibers, about 5% for the UHPC beams with longitudinal reinforcement and steel
fibers, and about 5% for UHPC beams with both longitudinal and shear reinforcement and without steel
fibers.
One can see clearly the differences between NSC and UHFRPC beams by comparing Cases B1, B3,
and B5 (Cases B4 and B6 of UHPC beams without steel fibers are not that practical, since the fibers have
a major role in achieving the enhanced behavior). Case B5 shows that traditional shear reinforcement is
very important also in UHFRPC beams, without which one will not be able to attain the enhanced
performance. Then, Cases B1 and B3 show that the difference in performance between NSC and
UHFRPC beams with the same reinforcement details is quite modest. This brings into question the merit
in replacing inexpensive NSC with a much more expensive UHFRPC for the same reinforcement detail.
Clearly, just replacing the concrete material could be unjustified, and one needs to address the combined
effects of the concrete material with the reinforcement detail. This effort is currently underway.
The behavior of NSC and UHPC slabs under static and shock tube loading tests were conducted
separately, and they are described in an accompanying paper. Nevertheless, that study contributed to
defining the compression membrane behavior, and to identify requirements for follow up studies to define
the transition into the tensile domain.
3.1.3 Direct shear behavior

Direct Shear is a sudden and catastrophic failure type commonly observed in reinforced concrete
structures under highly impulsive dynamic loads. It is a sliding type failure where slip occurs
perpendicular to the member’s axis at a location where the load or geometry is discontinuous. Static
and impact tests were conducted on NSC, UHPC and UHFRPC specimens with 0%, 0.8% and 1.6%
direct shear reinforcement, as shown in Fig. 6, and the data were analyzed in both the time and
frequency domains [33-36]. The resulting shear stress versus slip curves were compared to the
Hawkins shear model, and appropriate changes were proposed to the Hawkins shear model for both
NSC and UHPC cases, as shown in Table 3. However, these changes had only small effects on the
simulated behaviors of buried RC box-type structures [37]. Nevertheless, the analyses of the direct
shear behavior in the frequency domain have highlighted new details that enabled one to identify
more accurately the underlying behavioral parameters, as will be presented in an accompanying
paper.
(a) (b)
Fig. 6: Push-off specimen geometry, (a) specimen geometry, (b) resistance model
24
KEYNOTE PAPERS
Table 3: Direct shear resistance parameters

Original Hawkins NSC UHPC UHFRPC
1
165 0.157 ′ 165 0.157 ′ NA NA
8 ′ 0.8 11 ′ 0.6 11 ′ 1.3 11 ′ 1.3

0.9 `
0.85 ` 0.008 2.5 0.008 1.5
Ke 12 ′ NA NA
Kc 3.5 ′ 1.2 ′ 1.9 ′
Ku 2000 0.75 ′ 2000 0.75 ′
Δ1 0.004 /
Δ2 0.012 / / /
Δ3 0.024 0.012 0.012 0.033

0.008 0.008 0.008
Δ4 / / / /
Δmax 1
,
120
900
`
2.86
3.1.4 Columns
The behavior of NSC, UHPC and UHFRPC columns has also been studies numerically to better
understand the effects of using different concrete material mixes on the response to different levels of
explosive loading [38,39]. These analyses were performed in both the time and the load-impulse (P-I)
domains. It was noted that UHFRPC columns could survive four time as much impulse as NSC
columns with the same reinforcing details, and this highlighted the clear advantage of using
UHFRPC for specific applications.
4. Summary and Conclusions
Addressing both the load and structural resistance sides of the dynamic equation of equilibrium
is an underlying requirement for protective facilities to survive known explosive threats. Both
Thermodynamic and mechanical load suppression must be combined with advanced structural
resistance capabilities that are based on efficient combinations of materials and structural detailing.
Explosive loads must address the combined blast-fragment effects to provide a realistic basis for
analysis and design. Such treatment of the problem should be attempted with considerations of
energy flow into and through a facility, as described in an accompanying paper. UHFRPC materials
have been introduced for special construction, and they have the potential of enhancing the behavior
and performance of protected facilities. Recent studies, however, have highlighted that one may not
be able to proceed toward this objective by merely replacing NSC with UHFRPC. The Physical
phenomena that govern structural behavior have shown meaningful differences when such
substitution were examined, and this underlined the need for developing appropriate structural
behavioral models and design guidelines to be used for the desired applications.
25
KEYNOTE PAPERS
Acknowledgements
The generous support by the Defense Threat Reduction Agency (DTRA), the US Army
Engineer Research and Development Center (ERDC), and NATO that enabled the various studies
presented in this paper are gratefully acknowledged.
References
[1] US DOD, DoD Minimum Anti-Terrorism Standards for Buildings, UFC 4-010-01, Change 1, 1
October 2013.
[2] US DOD, Structures to Resist the Effects of Accidental Explosions, UFC 3-340-02, Change 2, 1
September 2014.
[3] Department of the Army, Fundamentals of Protective Design for Conventional Weapons,
Technical Manual TM 5-855-1, November 1986.
[4] US DOD, Design of Buildings to Resist Progressive Collapse, UFC 4-023-03, 14 July 2009.
[5] ASCE, Design of Blast Resistant Buildings in Petrochemical Facilities, 2nd ed., American
Society of Civil Engineers, 2010.
[6] ASCE, Blast Protection of Buildings, American Society of Civil Engineers, ASCE Standard,
2011.
[7] Dusenberry, D.O., Handbook for Blast Resistant Design of Buildings, Wiley, 2010.
[8] Krauthammer, T., Modern Protective Structures, CRC Press, 2008.
[9] Mehlin, H. P., and Parr, W., Fundamental Investigations on the Effects of Combined
Impulse of Blast and Fragments. Ernst Mach Institute, 1983.
[10] Koos, R., Die kombinierte Luftstoss-splitter-Last, eine quantitiv bestimmbare Grösse für die
Schutzbaubemessung, Internationales Symposium, Interaktion Konventioneller Munition mit
Schutzbauten, Band I, Mannheim, Germany, pp. 348-366, 9-13 Maertz 1987.
[11] McCarthy, G., and Krauthammer, T., Combined Blast and Fragment Loading Definition, 2nd
Design and Analysis of Protective Structures (DAPS 2006), Singapore, 13-15 November, 2006.
[12] Burchfield, C., French, R., Jaberi, B., Kim, J., Maher, E., Tsai, J., and Krauthammer, T.,
Assessing the Capabilities to Predict Combined Blast and Fragment Effects, 22nd International
Symposium on Military Aspects of Blast and Shock (MABS 22), Oxford, UK, September 8-12,
2014.
[13] Gelfand, B.E., and Silnikov, M.V., The Selection of the Effective Blast Reduction Method when
Detonating Explosives, J. Phys. IV France, Vol. 12, pp 371-374, 2002.
[14] Mataradze, E., Krauthammer, T., Chagelishvili, E., and Chikhradze, N., System for Protecting
People and Underground Facilities from Internal Terrorist Explosions, Proc. 2nd International
Conference on Design and Analysis of Protective Structures 2006 (DAPS2006), Singapore,
13-15 November 2006.
[15] Mataradze, E., Krauthammer, T., Chikhladze, N., Chagelishvili, E., and Kalichava, I., Choice of
Design Parameters of an Underground Shock Wave Absorber, Proc. 13th International
Symposium on the Interaction of the Effects of Munitions with Structures (ISIEMS 13), Bruehl,
Germany, 11-15 May 2009.
[16] Mataradze, E., Krauthammer, T., Chikhradze, N., and Chagelishvili, E., Influence of Liquid
Phase Concentration on Shock Wave Attenuation in Water Mist, Proc. 21st International
Symposium on Military Aspects of Blast and Shock, Jerusalem, Israel, 3-8 October, 2010.
[17] Chauvin, A., Zrbib, J., Jourdan, G., Daniel, E., Mariani, C., Houas, L., Biamino, L., Tosello, R.,
and Praguine, D., Investigation of the Attenuation of a Shock Wave Passing Through a Water
Spray, Proc. 21st International Symposium on Military Aspects of Blast and Shock, Jerusalem,
Israel, 3-8 October, 2010.
[18] van der Wal, R., Cargill, S., Longbottom, A., Rhijnsburger, M., and van Erkel, A., Explosion
Mitigation by Water Mist, Proc. 21st International Symposium on Military Aspects of Blast and
Shock, Jerusalem, Israel, 3-8 October, 2010.
[19] Mataradze, M., Krauthammer, T., Chikhradze, N., Chagelishvili, E., Bochorishvili, N., and
Akhvlediani, I., Designing of Water Mist Generator for Attenuation of Shock Wave in Tunnels,
Proc. 4th International Conference on Design and Analysis of Protective Structures (DAPS),
Jeju, South Korea,19-22 June 2012.
26
KEYNOTE PAPERS
[20] Mataradze, E., Krauthammer, T., Chikhradze, N., and Chagelishvili, E., Shock Energy Absorber
for Protection Underground Structure from Internal Explosions, Proc. 15th International
Symposium on the Interaction of the Effects of Munitions with Structures (ISIEMS 15),
Potsdam, Germany, 16-20 September 2013.
[21] Mashukov V. I., Blast Impact on the Environment, Moscow, Nedra, 1976 (in Russian).
[22] Gurin A.A., et al., Shock Waves in Mine Opening, Moscow, Nedra, 1983 (in Russian).
[23] Krauthammer, T., and Yim, H.C., Exploratory Collaborative Development of an Advanced
Load-transfer Limiter for Blast, Shock and Impact Protection, Technical Report to Idaho
National Laboratory, CIPPS-TR-003-2009, Center for Infrastructure Protection and Physical
Security, University of Florida, June 2009.
[24] Starr, C.M., and Krauthammer, T., Testing of Autoclaved Aerated Concrete Cladding Panels
under Impact Loads, Final Report to U.S. Army ERDC, PTC–TR-003-2003, Protective
Technology Center, Penn State University, August 2003.
[25] Astarlioglu, S., and Krauthammer, T., State-of-the-art Report on Fiber Reinforced Ultra-high
Performance Concrete, 2nd edition, Technical Report to DTRA, CIPPS–TR-001-2013, Center
for Infrastructure Protection and Physical Security, University of Florida, January 2013.
[26] Astarlioglu, S., Bui, L., and Krauthammer, T., The Behavior of Reinforced Concrete and
Ultra-high-performance Concrete Columns with Plastic Hinges under Impact and Blast
Loads,” Technical Report to DTRA, CIPPS–TR-002-2013, Center for Infrastructure Protection
and Physical Security, University of Florida, January 2013.
[27] Friedrich, N., Astarlioglu, S., and Krauthammer, T., Impact Testing of Normal-Strength and
Ultra-High-Performance Concrete Cylinders, Technical Report to DTRA, CIPPS–
TR-005-2013, Center for Infrastructure Protection and Physical Security, University of Florida,
January 2013.
[28] Stone, M., Koch, D., Krauthammer, T., Astarlioglu, S., and Bui, L., Impact Testing of
Normal-strength and Ultra-high-performance Concrete Beams, Technical Report to DTRA,
CIPPS–TR-006-2013, Center for Infrastructure Protection and Physical Security, University of
Florida, January 2013.
[29] Astarlioglu, S., Krauthammer, T., and Bui, L., Penetration Resistance of Normal-strength and
Ultra-high-performance Concrete, Technical Report to DTRA, CIPPS–TR-007-2013, Center
for Infrastructure Protection and Physical Security, University of Florida, September 2013.
[30] French, R., Stone, M., Krauthammer, T., Kim, J., Bui, L., and Lavenhagen, D., Assessment of
Direct Shear Behavior in Normal-strength and Ultra-high Performance Concretes, Technical
Report to DTRA, CIPPS–TR-008-2013, Center for Infrastructure Protection and Physical
Security, University of Florida, September 2013.
[31] Krauthammer, T., Astarlioglu, S., Bui, L., Koch, D., Stone, M., Lavenhagen, D., French, R., and
Friedrich, N., Normal-strength and Ultra-high-performance Concrete Behavior,” Technical
Report to DTRA, CIPPS–TR-009-2013, Center for Infrastructure Protection and Physical
Security, University of Florida, September 2013.
[32] Faust, B., Investigation of Compressive Membrane Action in Ultra High Performance Concrete
Slab Strips, Doctoral Dissertation, University of Florida, November 2013.
[33] French, R., Stone, M., Krauthammer, T., Kim, J., Bui, L., and Lavenhagen, D., Assessment of
Direct Shear Behavior in Normal-strength and Ultra-high Performance Concretes, Technical
Report to DTRA, CIPPS–TR-008-2013, Center for Infrastructure Protection and Physical Security,
University of Florida, September 2013.
[34] Kim, J., Krauthammer, T., French, R., and Stone, M., Frequency Domain Assessment of Direct
Shear in NSC and UHPC, Technical Report to DTRA, CIPPS–TR-001-2014, Center for
Infrastructure Protection and Physical Security, University of Florida, May 2014.
[35] French, R., Stone, M., Kim, J. and Krauthammer, T., Assessment of Direct Shear Behavior in
Normal and Ultra-high Performance Concretes, 22nd International Symposium on Military
Aspects of Blast and Shock (MABS 22), Oxford, UK, September 8-12 2014.
[36] Kim, J., Krauthammer, T., French, R., Stone, M., and Maher, E., Frequency Domain Assessment
of Direct Shear in NCS and UHPC, 22nd International Symposium on Military Aspects of Blast
and Shock (MABS 22), Oxford, UK, September 8-12 2014.
[37] Krauthammer, T., and Astarlioglu, S., “The Effects of Direct Shear Resistance Models on the
Simulated Behavior of Buried RC Roof Slabs Subjected to Airblast-induced Ground Shock,” 22nd
27
KEYNOTE PAPERS
International Symposium on Military Aspects of Blast and Shock (MABS 22), Oxford, UK,
September 8-12 2014.
[38] Astarlioglu, S., Bui, L., and Krauthammer, T., “The Behavior of Reinforced Concrete and
Ultra-high-performance Concrete Columns with Plastic Hinges under Impact and Blast Loads,”
Technical Report to DTRA, CIPPS–TR-002-2013, Center for Infrastructure Protection and
Physical Security, University of Florida, January 2013.
[39] Astarlioglu, S., Krauthammer, T., Bui, L., “Plastic Hinge Formation in NSC and UHPC Columns,”
Engineering Structures, Vol. 61, pp. 1-12, March 2014.
28
KEYNOTE PAPERS

0940 - 1020 hours
Numerical Studies on Reinforced Concrete Sub-Structures under

Progressive Collapse Caused By Contact Detonation
Anh Tuan Pham*, Kang Hai Tan
School of Civil and Environmental Engineering, Nanyang Technological University, Singapore

(atpham@ntu.edu.sg), (ckhtan@ntu.edu.sg)
Abstract
To study the effect of blast pressure on structural resistance against progressive collapse under
column removal scenario induced by contact detonation, and to investigate the development of
catenary action within dynamic regime, a physics-based finite element model is developed and
presented in this paper. The model is first validated by a blast-induced test on reinforced concrete
(RC) sub-assemblages under middle column loss assumption. The sub-assemblage included a
two-span beam, a middle column and two column stubs at both sides. After which, parametric studies
were conducted to investigate the influence of some important dynamic and structural factors such as
boundary stiffness, damping ratio, charge weight attached to the middle column, etc. The study shows
that catenary action in sub-assemblages can help to prevent a structure from collapse even when the
bottom longitudinal reinforcement in beams has already fractured. Moreover, stiffness of horizontal
restraints plays an important role to mitigate disproportionate collapse. Nonetheless, when beams
cooperate with slabs, the beam-slab system seems to be less influenced by this horizontal restraint
condition. A comparison was made between nonlinear dynamic procedure and nonlinear static
analysis incorporating simplified energy method for dynamic assessment. It was concluded that the
simplified static approach in lieu of dynamic analysis may underestimated the resistance of structures
to withstand collapse, especially under severe damage conditions. Thus, the simplified approach can
be considered as a conservative static method for practical design.
Keywords: Reinforced concrete; Progressive collapse; Contact detonation; Catenary action
1. Introduction
The risks of progressive collapse on governmental and public buildings are substantially
increased nowadays due to heightened danger for terrorist attacks. Several methods and design
guidelines have been released to help mitigate structures from progressive collapse. Among them,
direct method using Alternate Load Path (ALP) approach is an effective solution to investigate
structural resistance to progressive collapse[1, 2]. However, its main assumptions, i.e. the
single-column removal scenario, is criticized as an un-realistic situation for collapse induced by a
blast event[3] due to the neglect of initial damages from blast to structures when applying this
column-loss assumption.
Due to complexity and extensive resources required for nonlinear dynamic analysis, this
procedure is less preferred for investigating structural response under progressive collapse scenarios.
Instead, a linear or nonlinear static procedure incorporating equivalent dynamic factor is usually
adopted. Dynamic effects can be considered through load-increase factors [1, 2] or by using some
simplified methods based on energy balance [4]. Such kind of analysis is computationally efficient
but needs to be verified by actual dynamic tests.
Recently, there are extensive experimental studies on the ALP of RC structures [5-7]. Most of
them apply quasi-static approach to investigate structural responses against progressive collapse
29
KEYNOTE PAPERS
situations. The mobilization and development of both compressive arch action and catenary action,
which depend strongly on lateral restraint conditions, were clearly observed. While compressive arch
action is an efficient way to enhance the maximum bending capacity of the beam section, catenary
action can be considered as the last safety net to prevent the structure from complete collapse.
Nonetheless, the capacity of catenary action under dynamic regime is not yet confirmed
experimentally. To study the development of catenary action under both quasi-static and dynamic
regimes, and to quantify the effect of blast loads on RC structures caused by contact detonation,
numerical analyses based on physics-based finite element method (FEM) are developed and reported
in this paper. Results from the study confirm the enhancement of catenary action in preventing
collapse, which was not observed in tests [8]. Several important factors which significantly affect the
mobilization of catenary action are considered in this paper. The research concludes that the
simplified frame-work on dynamic assessment proposed by Izzuddin [4] is acceptable when the
combined effects of both damping and blast are neglected.
2. Numerical validations on beam-column substructures
2.1. Quasi-static and contact detonation tests on RC sub-assemblages under column removal
scenario
An experimental study on progressive collapse resistance of beam-column sub-assemblages

with the scenario of middle column removal was conducted in 2010 at Nanyang Technological
University [7]. The structure consisted of a two-span beam with a middle-joint and two column stubs.
Two RC specimens with the same geometry and boundary conditions, namely, S1 and S2 are tested.
The column removal scenario was simulated by gradually increasing the displacement of the
middle-joint (Fig. 1). Fig. 2(a) shows the 3-D setup of the tests whereas Fig. 2(b) presents overall
behaviour of the specimens. The test results clearly showed three phases of behaviour including
purely flexural action at the beginning, followed by compressive arch action and then catenary action.
The onset of catenary action was indicated by the change of horizontal reactions from compression to
tension and the structure failed when both the top and bottom longitudinal rebars in the two-span
beam fractured. According to test results, it was clearly shown that catenary action in beams could
still develop significantly even after the fracture of bottom rebars.
Fig. 1: Test setup[7]
(a) Test setup (b) Test results

Fig. 2: Quasi-static test [7]
30
KEYNOTE PAPERS
To investigate the effect of ultra-fast dynamics compared to the structural response from static
tests [7], a series of progressive collapse tests induced by contact detonation was conducted at
Fraunhofer EMI, Germany in 2012 [8]. The specimen design and boundary conditions were similar to
specimen S2 from the static tests conducted at NTU. Column removal was simulated by detonating a
C-4 charge placed at the middle column of the specimen. Specimen design and general configuration
of the contact detonation tests are shown in Fig. 3 and Fig. 4. Two test results, i.e. SD-2 and SD-3,
were used in this paper to validate proposed FEM models. These two specimens had similar design,
test setup and charge weight at the middle column. The additional applied loads on the middle
column, however, were 27 and 47 kN for SD-2 and SD-3, respectively.
Fig. 3: Specimen design [8]
Fig. 4: Test setup [8]
Test results of specimen SD-2 showed that, after the middle support was removed by blast
pressure, the middle joint experienced uplift within the first 100 ms before falling freely under gravity
load and finally sustaining a permanent deflection of 50.2 mm (Fig. 5(a)). SD-2 behaved in a flexural
manner with the appearance of compressive arch action (denoted by the mobilization of horizontal
forces at two end supports). Compared to SD-2, SD-3 withstood a larger additional load of 47 kN and
responded beyond the maximum compressive arch action capacity. Displacement of the middle joint
kept increasing until the specimen hit the ground at 471.7 mm (Fig. 8(b)). The horizontal reaction of
SD-3 was compression at the beginning and changed to tension when the middle joint deflected more
than one beam-depth, indicating the contribution of catenary action before failure occurred.
31
KEYNOTE PAPERS
(a) Specimen SD-2 (b) Specimen SD-3

Fig. 5: Test results[8]
2.2. Numerical model
In this part, numerical analyses were conducted to study the behaviour of RC beam-column
sub-assemblages against the scenario of supporting column loss under both quasi-static and blast
conditions. An explicit finite element software LS-Dyna [9] was used to simulate the
sub-assemblages due to its abilities in modelling the blast pressure on structures as well as a wide
variety of available constitutive models in the programme.
Based on symmetry assumption for the geometry, boundary conditions and material properties
from the tests, only one-haft of the specimen was modelled as shown in Fig. 6. Concrete was
simulated by8-node solid elements with reduced integration scheme. To avoid zero energy modes
(hour-glass modes) that might arise within the analysis procedure, the Flanagan-Belyschko stiffness
form was applied. Horizontal restraints at two sides of the specimens were modelled by two series of
lateral nonlinear elastic springs to take into account the difference in stiffness between tension and
compression. Reinforcing bars were explicitly modelled by the 2-node Hughes-Liu beam element
with 2x2 Gauss quadrature integration. This beam formulation is based on a degeneration of the
8-node solid element. Therefore, the Hughes-Liu beam element is compatible with brick elements,
and yet provides computational efficiency and robustness. The Hughes-Liu beam formulation
considers the behaviour of axial force, bi-axial bending, and includes finite transverse shear strains.
Lateral springs Add-load by solid plate with

Lateral springs
Displacement –controlled
equivalent mass
Symmetric
boundary Roller Extended middle column
(a) FEM model for quasi-static test (b) FEM model for blast test
Fig. 6: Symmetrical model in LS-Dyna
The continuous surface cap model MAT_159 was employed to simulate the behaviour of
concrete material. It can effectively capture post-peak softening, shear dilation, confinement effect,
and strain rate hardening. An isotropic elastic-plastic material model
Mat_Piecewise_Linear_Plasticity (MAT_024) was used for steel reinforcement which also
considered strain rate effect. The steel material was assumed to be identical in tension and
compression. The unloading path was assumed to occur elastically.
32
KEYNOTE PAPERS
Composite action between concrete and reinforcement was mobilized through bond stress-slip
relationship. The assumption of perfect bonding may over-estimate the bearing capacity, and leads to
premature fracture of rebars. In this numerical study, the bond-slip behaviour was modelled by a
one-dimensional contact function (Contact_1D) between the concrete solid elements and the rebar
beam elements. The slave nodes from beam elements were constrained to the master nodes from solid
elements through virtual springs. The value of the spring force depended on the relative displacement
between slave nodes and master nodes[9].
The bond-slip relationship proposed in CEB Model Code 2010[10] was applied herein. The
bond stress  between concrete and rebars for pull-out failure is listed as follows:
  / for 0 (1)
  for (2)
    / for
(3)
  for
(4)
The parameters  ,  , , , were defined following CEB 2010 and depended on the
compressive strength of concrete as well as the bonding condition. On the other hand, the bonding
relationship within Contact_1D (LS-Dyna) was assumed to be elastic-perfectly-plastic. After
attaining the maximum shear stress, the bond stress would decay following an exponent damage
curve. The constitutive law of bond stress-slip can be represented by:
 for (5)
  for (6)
where is the bond shear modulus, is the maximum elastic slip, is the damage curve
exponential coefficient and is the damage parameter. Fig. 7 shows the comparisons between the
CEB and the Contact-1D model for bond-slip.
Slip (mm)
16
14
Bond stress (MPa)
12
10
8
6
4 CEB model
2 Proposed model for Contact_1D
0
0 5 10 15
Fig. 7: Bond stress-slip relationship based on CEB model and simplified model for Contact-1D
The static test results[7]showed that the bond-slip behaviour and pull-out phenomena occurred
mostly within the connection regions. To save computational cost, the one-dimensional contact
function was applied for beam longitudinal rebars in these regions only. For longitudinal rebars in
other parts of the beams, perfect bonding between concrete and reinforcement was assumed. The
LS-Dyna keyword Constrained_ Lagrange_In_Solid was used to embed the rebar beam elements into
the concrete solid elements at these regions.
The additional applied load on the middle-column was simulated by an integrated solid plate
placed on top of the middle-joint. The material density of these special elements was adjusted to be
33
KEYNOTE PAPERS
equal to the weight of the external loading system. This type of modelling can take into account the
mass of the additional load together with the total mass of the structural system instead of considering
the imposed load as an ordinary load-case.
2.3. Simulation of blast pressure in LS-Dyna
In order to calculate the blast loads acting on the assemblage structure, a blast load function
integrated with LS-Dyna, Load_blast keyword, was implemented in this study. This function
employs analytical formulae and blast data available in the CONWEP programme[11]. It allows users
to define a specific weight of equivalent TNT explosion with a certain position and a detonation time
of the blast. Three main blast scenarios are available, namely, spherical air-burst, hemispherical
ground-burst, and spherical air-burst with ground reflection [9]. The time-history of blast loads will
then be calculated automatically and applied to the targeted surfaces of structural models. The value
of both incident and reflected pressure can be computed for a specific position. The applicable range
of this function is for the scaled distance from 0.147 to 40 m/kg1/3 for the spherical air-burst and from
0.178 to 40 m/kg1/3 for the hemispherical surface-burst. Based on the charge weight of the composite
explosive C-4 used in the test, an equivalent TNT charge weight was chosen to generate an equivalent
blast impulse. A value of 600 g ideal TNT was chosen for the Load_blast function used in both SD-2
and SD-3 simulation. The location of the blast was selected under the bottom of the middle-column to
simulate the effect of blast on the assemblage. Blast loads were assigned for the bottom faces and side
faces of both the beam and the middle-column. Since the blast effects were not measured in the actual
test, it was difficult to validate exactly its effect on the sub-structure. However, the blast loads on the
specimen could still be verified by comparing the general behaviour between the test results and the
FEM simulations. Moreover, a qualitative comparison of damage patterns between the tests and the
models could be made for the end part of the middle column.
2.4. Strain rate effect
In practical engineering, strain rate effects on material behaviour are usually represented by a
Dynamic Increase Factor (DIF) which is the ratio of dynamic strength to static strength with regard to
a specific strain rate. The CEB Model Code for Concrete Structures [10] proposes practical equations
to determine this DIF for compression and tension strength of concrete with the assumption that it
only depends on the strain rate. These equations are applied in this study to take into account the
effect of strain rate on concrete material. For reinforcing steel, the empirical equations on strain rate
enhancement factors for steel proposed by Malvar [12] was utilized and the DIF was determined for
both yield strength and ultimate strength.
2.5. Boundary conditions
For the quasi-static tests [7], the stiffness from horizontal restraints were carefully monitored
and such data were applied into the FEM model to validate test results. From the blast tests [8],
however, the reaction stiffness of horizontal restraints was not quantitatively measured. Hence, it was
considered as a variable that had to be determined by trial and error. The importance of this horizontal
restraint stiffness to the overall behaviour of RC sub-structures was clearly established from
literature. Based on a general understanding of structural dynamics, it is shown that the fundamental
period of a single-degree-of-freedom system mainly depends on mass and stiffness by the following
equation:
1  (7)
1 
Where wD and wn are the damped circular frequency and natural circular frequency of the system, k
and m are stiffness and mass of the system, respectively, and  is the equivalent damping ratio.
By assuming the fundamental mode of the sub-assemblage system under free-fall condition is
the vertical vibration of the middle joint, the fundamental period can be determined using equation
(7). From test results of SD-2, the fundamental period of vibration within the gravity-load phase was
310 ms. For an RC member under dynamic vibration, the damping ratio of the global system can be
34
KEYNOTE PAPERS
specified as 5% and it usually does not affect the value of the frequency. Moreover, the mass of the
sub-assemblage system was already known. Therefore, only the k-stiffness was a variable which
depended on different values of lateral spring stiffness. The stiffness values of top and bottom springs
were assumed to be equal. From trial and error, total lateral stiffness within the range of 3.5×104 to
5×104kN/m gave reasonable predictions of the fundamental period. In this study, the stiffness value of
3.5×104kN/m was chosen because it could provide the best fit with test data.
2.6. Analysis procedure and validation results for quasi-static tests
Initially, gravity load was gradually applied, and then the middle-joint was forcibly moved
downwards in a displacement-controlled manner until both bottom and top reinforcing bars of beam
fractured. Explicit analysis procedure was employed to ensure stability and convergence. The results
from numerical analysis are presented and compared to test results in this section. Generally, the
FEM models show reasonably good agreement with experimental data in terms of overall response,
horizontal reactions as well as failure modes of both concrete and reinforcement.
Fig. 8 shows the comparisons between test and numerical results with regard to middle joint
displacement versus applied load and horizontal reaction. The FEM models match well with testing
for overall behaviour of the structure. In both validations for applied load, numerical models predict
quite well the first peak from compressive arch action and the final period due to residual capacity
from catenary action. In terms of damage patterns, numerical models agree fairly well with the test
results for both concrete and longitudinal reinforcement. Fig. 9 shows the comparison of the final
stage of S1 between the test and the simulation. In the FEM results, concrete failure is represented by
the value of Damage index and the erosion of concrete element. An index of 0 denotes no failure
while an index of 1 indicates that concrete material is totally damaged. It is easy to notice that both
the concrete damage by cracking (at the beam surface near the end-joint and at the middle-span) and
by spalling (at the beam-end near the middle-joint) are successfully simulated in the model. Besides,
the fracture of bottom rebars at the middle-joint and top rebars at the end-joint are explicitly
simulated in the models.
Displcement (mm)
80 400
Applied load - Test
Applied load - FEM
60 Horizontal load - Test 300
Horizontal Reaction (kN)
Horizontal load - FEM

Applied load (KN)
40 200
20 100
0 0
0 100 200 300 400 500 600
-20 -100
-40 -200
Displcement (mm)
80 400
Applied load - Test
Applied load - FEM
60 300
Horizontal Reaction (kN)
Horizontal Reaction - Test

Horizontal reaction - FEM
Applied load (KN)
40 200
20 100
0 0
0 100 200 300 400 500 600 700
-20 -100
-40 -200
(a) S1 (b) S2
Fig. 8: Validation results of S1, S2
35
KEYNOTE PAPERS
(a) Photographs of final stage
(b) Damage index and concrete erosion from FEM model
Fig. 9: Comparison of damage partterns of S1
2.7. Analysis procedure and validation results for contact detonation tests
At the beginning of the analysis, the explosive defined by the Load_blast function was activated
and only lasted for a few milliseconds. Due to the uplift pressure from the blast loading, the
assemblage started to deform upwards. The additional gravity load on top of the middle joint was
then applied at 80 ms when the sub-structure achieved the maximum upward displacement.
Thereafter, the sub-structure started oscillating and eventually all movements stopped due to inherent
damping effects. In this study, the global damping coefficient of the numerical model was set to be
5% of critical damping as recommended for general RC structures under earthquake conditions.
The numerical analyses of SD-2 and SD-3 were performed and the results are presented herein.
In general, the deformation time-history of the middle joint has been successfully validated in both
FEM models. In terms of horizontal reactions, the numerical results reasonably agree with the test
data, although some divergence in value is observed (Fig. 10, Fig. 11). The results from FEM analysis
also confirm the method of blast load application used in LS-Dyna software. With regard to failure
modes, the predicted results from FEM are in good agreement with those from the tests (
Fig. 12). Some qualitative comparisons for damage patterns at the bottom-end of the middle
column also confirm the validation of charge weight as well as the method of applying the blast loads
(Fig. 13).
100 Time (ms)
Time (ms) Top reaction - Test
40
80
SD2 - Test Top reaction - FEM
60
Reaction force (kN)
20 Bottom reaction - Test

SD2 - FEM 40 Bottom reaction - FEM
Displacement (mm)
0
- 100 200 300 400 500 600 700 800 900 1,000 20
-20 0
-20 0 100 200 300 400 500 600 700 800 900 1000
-40
-40
-60 -60
-80
-80
-100
(a) Time history of middle joint displacement (b) Time history of horizontal reaction
Fig. 10: Validation results of SD-2
36
KEYNOTE PAPERS
100 Time (ms)

100 Time (ms)
0 SD3 - Test
50
Displacement (mm)
- 100 200 300 400 500 600
Reaction force (kN)

SD3 - FEM
-100
0
-200 - 200 400 600
-50
-300
SD3 - Test
-400 SD3 - FEM -100
-500 -150
(a) Time history of middle joint displacement (b) Time history of horizontal reaction
Fig. 11: Validation results of SD-3
(a) Final failures of SD-3
(a) Damage patterns of FEM model at 500ms
Fig. 12: Comparison of damage partterns of SD-3
37
KEYNOTE PAPERS
(a) Test results of SD-2 and SD-3 (same charge weight) (b) FEM results (same charge weight, different mesh)
Fig. 13: Comparison of damage patterns of middle-column
The maximum strain rates in longitudinal rebars of both SD-2 and SD-3 fell in the range of 0.2 to
0.3/s for the first 20ms of the simulation, which was within the effective time of detonation (see Fig.
14, Fig. 15). This result agrees well with the findings from the test results which concluded that the
effect of strain rate enhancement on rebar strength is not significant for the case of assemblages under
progressive collapse initiated by close-in blast [8].
(a) Test result, maximum = 0.3~0.4/s (b) FEM result, maximum = 0.28/s
Fig. 14: Strain rate of reinforcement at middle joint interface
(a) Test result, maximum = 0.2/s (b) FEM result, maximum = 0.22/s
Fig. 15: Strain rate of reinforcement at end joint interface
38
KEYNOTE PAPERS
3. Parametric studies on RC sub-structures under progressive collapse
Based on the validated models from both static and contact detonation tests, parametric studies
were conducted and the results are presented herein. Firstly, factors such as span-to-depth ratio and
horizontal stiffness were quasi-statically investigated to study their influence within the static regime.
Factors such as restraint stiffness, load level, damping ratio, and charge weight were dynamically
considered. Additionally, the energy approach for dynamic assessment [4]was also compared with
the numerical results with regard to two design situations, viz. free-fall tests as well as close-in
contact detonation.
3.1. Parametric studies under quasi-static conditions
It is shown that catenary action is strongly affected by the span-to-depth ratio of the original
beam before the column being removed ( / ). This upper-bound mechanism can only
mobilize efficiently if the factor is not very small. If this value is smaller than 7, then after the first
peak, structural response will decrease quickly and the enhancement of tensile force has almost no
influence (Fig. 16(a)). In such cases, flexural and compressive arch mechanisms will govern the
response and shear deformation is remarkable. The structure can even fail by shear prior to ductile
failure by flexural action.
To study the effect of axial restraint, a normalized stiffness factor was defined based on the
product of horizontal stiffness and beam span, divided by the cross-sectional area and elastic
modulus. Five values of were compared and the results presented in Fig. 16(b) clearly show that
has significant effect on the response of structure as well as the development of catenary mechanism
in beams. When the value of 2, it can be considered as perfectly rigid condition. On the other
hand, if the horizontal restraint is too weak, say < 0.07, catenary mechanism cannot properly
mobilize either. In this instance, the end joint can be considered as a purely rotational restraint.
Observing from Fig. 16(b), the compressive arch action capacity and the first development of
catenary action before bottom rebars fracturing are greatly dependent on the horizontal stiffness ratio
. The final catenary action capacity and the maximum displacement, on the other hand, are less
affected by this factor. That is to say, secondary catenary action, developing after the fracture of
bottom rebars, can effectively enhance the ultimate capacity of structures even with a relatively small
value of horizontal stiffness.
Displacement 80 Displacement
160
k1 = 5
140 k1 = 7 70
k1 = 9
120 60
Applied load (kN)
k1 = 11
Applied load (kN)
100 50
80 40
60 30 k2 = 2.271
k2 = 0.227
40 20 k2 = 0.045
20 10 k2 = 0.023
k2 = 0.011
0 0
0 100 200 300 400 500 600 700 800 50 150 250 350 450 550 650
(a) Different span-to-depth ratios (b) Different horizontal stiffness
Fig. 16: Investigation of catenary action mobilization
3.2. Effect of horizontal restraints under dynamic condition
Lateral reaction stiffness also plays a vital importance on the overall behaviour of assemblage
systems under the blast duration as well as under the free-fall vibration. Parametric studies were
conducted hereinto study the influences of lateral stiffness. Several stiffness values were chosen for a
39
KEYNOTE PAPERS
free-fall dynamic analysis of SD-2 model under a constant applied load of 27 kN. Results show that
with ideally rigid constraint, the maximum deflection could be reduced by 64% compared to the
stiffness value of 3.5×104kN/m which was chosen for the validation of the blast tests (Fig. 17).
Time (ms)
0
0 100 200 300 400 500 600 700 800
-10
Rigid
Displacement (mm)
-20 200000 kN/m

100000 kN/m
-30
50000 kN/m
-40 35000 kN/m
-50
-60
Fig. 17: Comparison of different lateral stiffness values
3.3. Charge weight
Different values of equivalent TNT charge weight were used to study the effect of initial blast
pressure to the overall behaviour and resistance of sub-assemblage structures. The charge weight
varied from 0.6 kg, similar to the actual test, to 5 kg of TNT explosive. The additional load was set to
27 kN which was equal to the load level of SD-2 test. A numerical model of free-fall dynamic test
with only gravity-load is also presented as a baseline for comparison. The results clearly show a
significant increase in displacement between the free-fall case and cases with contact detonation as
shown in Fig.18. Compared to the free-fall case, the model using 5 kg of TNT detonation achieved
67% greater displacement. This shows that initial effects of blast can cause a remarkable reduction in
load-bearing capacity of structural system under progressive collapse event. This kind of effect is not
taken into account when assessing the dynamic effects according to DOD [2]as well as the energy
approach by Izzuddin [4]. This notable finding confirms the significance of blast pressure initiated by
close-in detonation.
Time (ms)
40
20
Displacement (mm)
0
0 100 200 300 400 500 600 700 800
-20
Free-fall dynamic
-40
Contact detonation of 0.6kg TNT
-60 Contact detonation of 2.7kg TNT
-80 Contact detonation of 5kg TNT
-100
Fig.18: Effects of charge weight
3.4. Damping ratio
Different levels of damping ratio were used for the FEM model of SD-2 with other parameters
such as boundary stiffness, charge weight and load level kept unchanged. The damping ratios
were3%, 5% and 7% of critical damping, representing three different levels of concrete damage from
minor cracks, moderate damage to severe damage. The result from 0% damping was used as a
baseline for comparison. The computational results clearly show that damping effect not only
contributes to the decay of vibration, but also reduces the maximum displacement, as well as the
magnitude of vibration (see Fig. 19). Within the first 100 ms (the blast-load phase), damping effect has
almost no influence on the general response of the structure. The energy dissipation from damping is
40
KEYNOTE PAPERS
not remarkable and can be neglected due to the short duration of blast loads. Nonetheless, it has a
dominant influence during the gravity-load phase. Hence, it can be concluded that the damping effect
on RC structures can produce a noticeable effect in mitigating progressive collapse, especially when
structures undergo extreme damages such as concrete crushing, rebars yielding and fracturing.
Time (ms)
20
0
0 100 200 300 400 500 600 700 800 900 1000
Displacement (mm)
-20
-40
-60
Damping ratio 7%
-80 Damping ratio 5%
Damping ratio 3%
-100 Damping ratio 0%
-120
Fig. 19: Effects of damping ratio
3.5. Comparison between dynamic analysis and static analysis associated with the energy approach
for dynamic evaluation
The pseudo-static method by Izzuddin [4] is computationally efficient and can predict the
column-removal scenarios as proposed by the DOD [2]. It is based on the simplified consideration
that all the work-done by gravity loads is equal to the strain energy absorbed in the structure. Based
on this, an example of comparison between two analysis procedures, namely, nonlinear static
associated with Izzuddin’s method and fully nonlinear dynamic, is made to verify the validity of
nonlinear static analysis as a substitute of the dynamic analysis. A numerical model of SD-2 specimen
with similar geometries, material properties and boundary stiffness to the actual test was analysed
under quasi-static procedure to determine its ultimate static capacity. Subsequently, the Izzuddin’s
energy method was applied to transfer the nonlinear static response of the FEM model to a
pseudo-static response to represent the dynamic effect. This approach uses the energy conservation
principle to convert dynamic response to static resistance, considering the sudden removal of column
as the sudden application of additional loading on the structure. At the moment when the structure
attains its maximum response, the velocity reduces to zero and the work-done by the external load is
equal to the strain energy of the structure. The maximum dynamic response can be calculated by
multiplying the static behaviour of a similar loading value with an amplifying factor b. For a given
static response, at each displacement-point (ud1, P0), the corresponding dynamic load-level Pn which
produces the same maximum displacement ud1 as the static load P0 can be determined as follows:
, (8)
,
Each point of this pseudo-static curve is specified by applying equation (8).
Simultaneously, fully nonlinear dynamic procedures were employed to assess the realistic
dynamic behaviour of the model under the sudden-loss scenario of middle column. Three different
cases of dynamic analysis are considered including free-fall dynamic without damping effect,
free-fall dynamic with damping ratio of 5%, and dynamic analysis using a contact detonation of 0.6
kg TNT without damping effect. For each dynamic case, the applied load on the middle joint was
varied from a small value to the greatest value that caused the structure to fail. At each load level, the
maximum displacement of the system was obtained and used to construct the dynamic response
curves. Comparisons of different cases are presented in Fig. 20.
41
KEYNOTE PAPERS
100 Displcement (mm)
90
fracture of bottom bars
80
70
Applied load (kN)
60
50
40
Quasi-static FEM
30
Pseudo-static response (Izzuddin)
20 Free-fall dynamic FEM (no damping)
Free-fall dynamic FEM (damping 5%)
10
Dynamic with 0.6kg TNT FEM (no damping)
0
0 100 200 300 400 500 600 700 800
Fig. 20: Comparison of different analysis procedures
It is clearly shown that results from the pseudo-static response match well with the results from
the analysis of free-fall dynamic analysis without damping and blast loads. Hence, the Izzuddin’s
energy method is reasonably suitable for the analyses of structures without considering damping and
blast effects.
For the case of free-fall dynamic with 5% of critical damping, the difference in maximum
displacement is observed when compared with zero damping. This divergence becomes remarkable
when the structure starts yielding, especially after the catenary action dominates the overall
behaviour of the system. This phenomenon is explained by energy dissipation from viscous damping
which can significantly reduce the effect of dynamic response. However, this ability can only be
effectively mobilized when structures have already experienced tremendous damages.
For the dynamic response associated with an initial blast pressure, noticeable increase in
maximum displacement is observed compared to the free-fall case. However, this reduction in
structural capacity is only significant when the applied load is less than 40 kN and the structure is
within the range of flexure or compressive arch action. As the applied load is increased and catenary
action is mobilized, the difference in maximum displacement between the case without initial blast
and the case with initial blast is almost unnoticeable. Further investigation on blast effects with larger
amount of detonation is necessary to fully understand the relationship between blast pressure and the
level of applied load.
For all three cases of dynamic analysis, when the applied load is relatively large, the bottom
reinforcing bars will fracture under dynamic response. However, the structure can still withstand the
final collapse based on the remaining top rebars. This observation confirms the capacity of catenary
action to mitigate final collapse even when the bottom rebars have already fractured. This
observation was confirmed in the static tests by Yu and Tan [7]. In the actual blast test of SD-3, the
testing procedure terminated when the specimen hit the ground with a corresponding displacement of
471.7 mm and bottom rebars had already fractured at one side of the middle joint. Hence, it is worth
noting that the ultimate catenary capacity of the assemblage was not confirmed in the test. The
specimen could still withstand the applied load of 47 kN if there was adequate headroom for vertical
deflection.
3.6. Effect of horizontal restraints in beam-slab system
Although the horizontal restraint condition plays an important role in the development of
catenary action for skeleton RC frames, its effect on tensile membrane mechanism in beam-slab
systems seems not to be obvious. In this study, a typical beam-slab system under a scenario of middle
column loss was chosen to investigate this effect. The system had similar span in both direction and
also had equal ratio for top and bottom reinforcement in slab. All beams were designed with identical
section and rebar arrangement. When a middle column is removed, it will affect all the four slabs
42
KEYNOTE PAPERS
which are supported by this vertical component. However, due to symmetry properties, only a quarter
of the affected area was modelled with an extension of the adjacent bay to simulate the rotational
restraint condition (Fig. 21(a)). Horizontal springs were arranged along the edge of the modelled area
to represent axial stiffness of the remaining structure to the studied area (Fig. 21(b)). Several values of
spring stiffness were chosen, which were zero stiffness, perfectly rigid, and 5*104kN/m. The notional
removal of the middle column was simulated by the gradual downward movement of this column,
representing the scenario of concentrated loading condition on slab.
Modelled area
Removed
column
P
Affected area
(a) Prototype building and investigated area (b) Quarter-type model

Fig. 21: FEM model for beam-slab analysis
Failure modes of all cases were similar, which included the fracture of bottom rebars in the
interior beams at the interface with the removed column together with punching shear failure at the
middle of the affected slab (Fig. 22(a)). That is to say, the behaviour of beam-slab system against
middle column removal under concentrated loading condition is not benefited by catenary action
from the top rebar of interior beams due to its early failure by punching shear. In terms of horizontal
restraints, the study showed that it only affect the first peak of compressive membrane action and
almost has no influence in the subsequent part of tensile membrane action (Fig. 22(b)). Therefore,
unlike the sub-assemblage system, the beam-slab system is somehow less influenced by the stiffness
of the horizontal restraint. In fact, the more importance boundary condition which significantly
affects the response of this system is the rotational restraint at the slab’s edge.
(a) Failure of beam-slab system (b) Effect of horizontal restraint

Fig. 22: Investigation results
43
KEYNOTE PAPERS
4. Conclusions
The LS-Dyna FEM models reasonably validated the quasi-static tests in terms of
load-displacement relationship as well as the actual blast tests in terms of displacement time-history
and effects of blast on the structure. For the blast test, nonetheless, it was more difficult to verify the
horizontal reactions due to the lack of test data on reaction stiffness. Both the test and the FEM results
show that strain rate enhancement does not have much influence on the overall behaviour of the
structure. The importance of span-to-depth ratio and horizontal restraint to the mobilization of
catenary action in RC sub-assemblages is remarkably highlighted in the study. Catenary action can
only be considered as significance if the span-to-depth ratio is not less than 7. Similarly, catenary
action also cannot produce any enhancement in structural capacity after the fracture of bottom rebars
if the normalized horizontal stiffness is smaller than 0.07. However, when beams are integrated with
slabs to behave as a unique system, the horizontal restraint has very limited influence in the overall
behaviour, especially the development of tensile membrane action in slabs. The studies also verified
the prediction of dynamic response using the energy method proposed by Izzuddin [4] for the case of
free-fall dynamic analysis without damping effect and blast loads. However, damping effect has an
indispensable influence on the behaviour of concrete structures. Noticeable reduction in strength due
to blast loads was also observed from the models. From the parametric studies, catenary action plays
an important role to resist collapse, even after the fracturing of bottom rebars. This conclusion needs
to be confirmed by actual dynamic and blast tests.
Acknowledgement
The authors gratefully acknowledge the funding entitled as “Development of a Design Guideline
and Analytical Tool to Mitigate Progressive Collapse of Buildings against Explosive Effects”, which
is provided by Ministry of Home Affairs, Singapore.
References
[1] General Services Administration (GSA). Progressive collapse analysis and design
guidelines for new federal office buildings and major modernization projects, Washington,
D.C, 2003.
[2] Department of Defense (DOD). Design of buildings to resist progressive collapse, Unified
Facilities Criteria (UFC) 4-023-03. Washington, D.C, 2013.
[3] Krauthammer, T. and H. Yim. Localized Damage Effects on Building Robustness. in
Structures Congress 2009@ Don’t Mess with Structural Engineers: Expanding Our Role.
2009. ASCE.
[4] Izzuddin, B., A. Vlassis, A. Elghazouli, and D. Nethercot, Progressive collapse of
multi-storey buildings due to sudden column loss—Part I: Simplified assessment framework.
Engineering Structures, 30(5): p. 1308-1318, 2008.
[5] Yi, W.-J., Q.-F. He, Y. Xiao, and S.K. Kunnath, Experimental study on progressive
collapse-resistant behavior of reinforced concrete frame structures. ACI Structural Journal,
105(4), 2008.
[6] Sadek, F., J.A. Main, H.S. Lew, and Y. Bao, Testing and Analysis of Steel and Concrete
Beam-Column Assemblies under a Column Removal Scenario. Journal of Structural
Engineering, 137(9): p. 881-892, 2011.
[7] Yu, J. and K.-H. Tan, Experimental and numerical investigation on progressive collapse
resistance of reinforced concrete beam column sub-assemblages. Engineering Structures,
55(0): p. 90-106, 2013.
[8] Yu, J., T. Rinder, A. Stolz, K.-H. Tan, and W. Riedel, Dynamic Progressive Collapse of an
RC Assemblage Induced by Contact Detonation. Journal of Structural Engineering, 140(6),
2014.
[9] Hallquist, J.O., LS-DYNA Keyword User’s Manual Version 971. Livermore Software
Technology Corporation, 2007.
44
KEYNOTE PAPERS
[10] CEB-FIP Model Code. Design of concrete structures, Fédération Internationale du Béton
fib/International Federation for Structural Concrete, 2010.
[11] Hyde, D., ConWep, conventional weapons effects program. US Army Engineer Waterways
Experiment Station, USA, 1991.
[12] Malvar, L.J., Review of static and dynamic properties of steel reinforcing bars. ACI
Materials Journal, 95(5), 1998.
45
KEYNOTE PAPERS

0900 - 0940 hours
Analysis and Design for Conventional Structures under Extreme

Multi-hazard Loading Conditions
Phillip L. Gould, PhD., P.E., S.E, F.SEI, Dist. M.ASCE
Washington University and St. Louis University, USA

(pgoul@seas.wustl.edu)
The essential characteristics of earthquake, blast and tornado loadings as they affect the capacity
of conventional structures are briefly described. For each loading case, the impact of a significant
event on a substantial structure is used to illustrate selected aspects of the structural performance. The
response to these events and their consequences are reviewed, similarities to and differences from the
requirements on popular lateral load-resisting system requirements are discussed, and multi-hazard
commonality characteristics are suggested. Also the application of contemporary nonlinear analysis
techniques is presented.
Keywords: Impact; Multi-hazard, Earthquake, Blast, Tornado, Pushover
1. Introduction
For conventional structures, as differentiated from nuclear and military facilities that present
unacceptable hazards from failure, the consideration of low probability high consequence loading
conditions has increased in the U.S. in recent years. While the very high price of recovery from recent
disasters may be a rational reason, this trend is also attributable to public reaction to some high profile
rare events such as the bombing of the Alfred P. Murrah Federal Building in Oklahoma, OK (1995)
and the tornados in Joplin, Missouri (2011) and Moore, OK(2014). Although quantitative risk
analysis might dismiss such events as too improbable to consider, the loading conditions associated
with them are increasingly being included in the structural analysis and design of selected
conventional structures.
Straight-line wind and earthquake loading have been addressed for decades in structural
engineering. As such, it is logical to examine the effectiveness of the structural systems provided to
resist these primarily lateral loads to also resist blast and tornado loads and to adapt some of the
familiar analysis techniques to deal with these additional extreme loads.
In this paper, the essential elements of earthquake, blast and tornado loadings as they affect the
capacity of conventional structures are briefly described. The impact of a significant unanticipated
event on a substantial structure is used to illustrate some aspects of structural performance under each
extreme loading. The response to these events and their consequences are reviewed with the objective
of identifying commonalities and differences in the performances and identifying critical elements
that contributed to the failures. In addition, general configuration and detailing practices that
contribute to good performance are suggested and nonlinear analysis techniques that may be
applicable to more than one hazard are discussed.
2. Extreme Lateral Loading Conditions
2.1 General
46
KEYNOTE PAPERS
The most common natural lateral loading conditions are straight-line wind, earthquake, rotating
wind and blast. While the first two are considered routinely in modern building codes, the latter two
are relatively unfamiliar to the designers of conventional structures. It is important to note that
earthquake loading was not always considered in some regions until rather recently so that an existing
structure that appears to be serviceable may be inadequately designed for current seismic provisions.
In this paper it is assumed that the straight-line wind conditions are the basis for the initial design of
the lateral load-resisting system and the extreme loads are the later three, earthquake, rotating wind
and blast.
The price paid for considering any lateral loading may be viewed simplistically as the increase in
cost of the superstructure, cladding and foundations over what is required for gravity loading. Also it
is important to recall that even if a structure was designed for the appropriate earthquake and wind
loading at the time, requirements for both of these load conditions have typically increased in
magnitude and complexity in many regions. For the purposes of this paper however, the straight-line
wind capacity of the system is assumed to be adequate.
2.2 Earthquake
Excitations from earthquakes are generally considered as delivered to the foundation of a

structure through the transmission of seismic waves from a source some distance away and beneath
the surface. There are other mechanisms such as extreme near field loading, liquefaction, landslides
and tsunami waves that may exacerbate the level of damage. Since the dynamic loading may be
applied over a relatively long duration, the maximum effects on the structure may be delayed and the
structural properties that affect the response modified during the application. As the dynamic loading
is transmitted from the foundation up through the superstructure, each component may be excited at
the system level and at the element level. The nature of the response is dependent on the dynamic
properties of the system or component as compared to those of the input waves. While the simple
mechanical concept of resonance is useful for visualizing the response of structures and their
components to particular earthquakes, it is insufficient due to multimode participation, damping and
inelastic action. Also, because the entire structure is subject to the loading, potential weaknesses may
be exposed in members that are not part of the primary lateral force-resisting system.
An illustration of the “weakness finding” characteristic of earthquake excitation is shown in
Figs. 1 where two practically identical chimneys are shown before and after the 1999 Tupras
earthquake in Turkey. The taller chimney on the right is 115m high with bottom diameter 10.3m, top
diameter 6.6 m and thickness varying from 45 to 20cm; its neighbor is similar but a few meters
shorter. The taller chimney has a 3.7m by 4.95m opening for a large heater pipe circumscribing an
angle of over 50o, which obviously interrupts the natural flow of vertical stresses and reduces the
circumferential rigidity but was within practice limits. The shorter chimney has no such opening.
The failure of the taller chimney shown on Figure 1(b) occurred at the large opening for the
heater pipe shown on Fig. 1(a), as discussed later in this paper [1].
Fig. 1: (a) Tupras Chimneys before earthquake Fig. 1: (b) Chimneys after earthquake
47
KEYNOTE PAPERS
2.3 Blast
Loading due to blast may originate from either accidental or man-made explosions. Blast
loading is often viewed as a nearly instantaneous peak overpressure followed by a slower decay to a
negative pressure. A graphical example is shown in Fig. 2. The incident shock waves may impact all
exposed surfaces of a structure, which in turn may generate an even larger reflected pressure pulse.
While blast loading resulting from external denotation may be visualized as an applied pressure over
a relatively small area of the exterior surface of the structure, the effects may propagate throughout
the structure due to damage to vital components of the primary structural system, especially the
primary frame members.
Fig. 2: Pressure Decay with Distance (FEMA 427)
The major distinctions between blast loading and other extreme loadings are summarized as [2]:
 The intensity of the localized pressures acting on building components can be several orders of
magnitude greater than those from other hazards.
 Explosive pressures decay extremely rapidly with distance from the source.
 The duration of the event is very short, measured in thousandths of a second, (milliseconds).
Similar to seismic loads, correctly representing the ductility and inelastic response of the different
building materials is critical to understanding the overall response of the structure. It has been
conjectured by an experienced forensic engineer that if the reinforced concrete Murrah Federal
Building had been detailed to the requirements for high seismic zones, the collapse may have been
avoided [3]. While such detailing was not required in that case, this observation emphasizes the
importance of locally strong and ductile components in collapse prevention.
Understandably the consideration of blast effects in designing conventional structures has increased
since the Murrah Federal Building bombing in 1995 in the US and similar events in other countries.
This led to analysis procedures initiated by the removal of one or more exterior columns and subsequent
steps for the evaluation of the progressive collapse potential of existing and new structures. The explicit
consideration of column removal, also referred to as the alternate load path method for progressive
collapse analysis, has now become a threat independent structural design procedure that is widely
applied for important structures. Obviously such approaches do not replicate the reflected pressure
48
KEYNOTE PAPERS
pulse effect shown in Fig. 2 or any specific loading scenario.

The blast failure shown in Fig. 3 occurred at the DeBruce elevator in Kansas in 1998, described
as the world’s largest grain elevator. The report [4] illustrates both the local impact of the explosion
and the propagation of the shock waves through connected confined spaces. The elevator was laid out
in a northeast/southwest direction but, for purposes of easy reference, it was generally described as
though it was oriented north/south. Fig. 3(a) is an aerial view taken of the east side of the elevator
within an hour after the explosion. The complex consisted of 246 circular grain silos that were 30ft. in
diameter and 120ft. high arranged in a linear array of 3 silos abreast. The 164 star-shaped spaces
between the circular silos, known as interstice silos, were also used for grain storage. There were
therefore a total of 310 grain storage silos in the elevator. Located midway between these storage
silos and separating them into a north array and a south array was a headhouse, 216.5ft. high from its
basement floor and standing 197ft. above ground. This structure was 42ft. square in cross-section and
housed four elevator legs as well as facilities to weigh and distribute grain into selected silos. The
overall length of the elevator, the headhouse and the silos, was 2,717 ft.
Fig. 3(a): DeBruce Grain Elevator viewed from East Side
Fig. 3(b): Close-up of Headhouse Fig. 3(c): Silos with Tops Blown Off
Beginning in the east tunnel of the south array of silos, a series of explosions following a crossover
tunnel propagated both directions in the two tunnels of the south array. Blast and fire blew upward
through the headhouse, Fig. 3(b), and into the south gallery from the headhouse for only a short
distance. Because the grain dust in that gallery near the headhouse had just been cleaned, the blast wave
separated from the trailing fireball causing the latter to self-extinguish. Although most of the remaining
south gallery remained integral, there was apparently more than ample unignited grain dust available to
have destroyed the remainder of the south gallery.
49
KEYNOTE PAPERS
The same blast wave and fire also moved out from the headhouse into the north gallery where it
continued while also diverting downward through empty silos into the west north tunnel. As blast waves
passed northbound beneath silos in both of the north array tunnels, they rose vertically through those
silos to the north gallery and blew off many silo tops as shown in Fig. 3(c). It is not likely that this
loading was anticipated in the design of the silo tops.
2.4 Tornado
As compared to the previous two loading cases of earthquake and blast, the engineering
understanding of tornado loading on buildings is at a significantly less sophisticated stage in the opinion
of the author. This is particularly true for mid and high-rise structures and those that are relatively large
in plan. A basic approach is to use the traditional energy equation to convert a velocity associated with a
maximum wind speed to a corresponding pressure and also to consider suction through an atmospheric
pressure drop. The maximum wind speed consists of a rotational and a translational component and it
has been suggested that importance factors, gust factors and variation of wind speed with height need
not apply. The later exception is significant and is not verifiable for taller structures. Another feature of
tornado loading is based on the Rankine vortex, an idealized axisymmetric vortex model. The wind
speed displays an increase from the center of the tornado outward to the Radius of Maximum Wind
(RMW) and a hyperbolic decay thereafter based on the decay exponent [5]. This indicates that the
maximum velocity pressure may occur only over a relatively small region of the structure in the plan
dimension. Similar to seismic and blast, designs that consider tornadic wind loads must focus on both
local and global load path issues within the structure.
Fig. 4(a): Tornado in Joplin, MO Fig. 4(b): Damaged Hospital: Note Intact Structural System
Several major structures including schools, gymnasiums, “big box” stores, and Mercy Hospital
within the St. John's Regional Medical Center in Joplin, MO took direct hits from a severe tornado Fig.
4(a) that struck the city May 22, 2011. In all there was perhaps $2 billion damage, 160 deaths and 1150
injuries. To the author’s knowledge, a direct hit of a severe tornado on a tall modern structure with an
engineered lateral load-resisting system, such as the hospital in Fig. 4(b), is exceedingly rare and the
resulting damage provides valuable new information on the potential destructive capability of tornado
loading. The tornado was classified as ES4 or ES5 and estimated as ½ to ¾ mile wide. As reported in
the media, damage was widespread and catastrophic at and around the building. The hospital lost many
windows, interior walls, ceilings, and part of its roof, and its life flight helicopter was also blown away
and destroyed. Six fatalities were recorded there and the nine-story building was so severely damaged
that it was later torn down. It was also reported that an engineering survey of the building revealed that
foundation and underpinning system were compromised.
According to multiple sources, such extreme structural damage to such a large and relatively
well-built structure was likely indicative of winds approaching 200 MPH. However, it is remarkable
that many elements of the lateral load-resisting system shown on Fig. 4(b) appear to be largely intact.
As reported in [5], the failure of the building envelope at St. John’s Regional Medical Center led to the
loss of protection and subsequent extensive damage to building interiors, including the electrical
distribution, HVAC and elevator systems. The envelope failure was the primary cause of the complete
50
KEYNOTE PAPERS
loss of functionality of this critical facility, which occurred despite the robust structural system that
withstood the tornado without structural collapse.
Because of the extensive damage, a new facility was designed with tornado resistance included in
the loading requirements. Some of the features include [6]:
 "Hard" exterior skin.
 Increased strength and stiffness of windows and glass curtain wall systems.
 Concrete-filled decks at selected roof areas to provide additional uplift resistance and
hardening.
 Increased column depth for moment-resisting lateral frames to improve overall building
stiffness.
 SidePlate connections for moment resisting lateral frames used to speed erection time and
reduce erection costs.
This type of structural system became popular following the Northridge earthquake.
3. Strength and Toughness of Structural Components
3.1 General
While the examples in the previous section and many other structural types achieve resistance to
lateral loading with different systems, there are some common characteristics. As suggested in [7],
primary areas that are common for all structural systems that resist extreme loading are capacity design,
ductile detailing and continuity. To these may be added transfer mechanisms to alternate load paths.
While all of these areas are interdependent and overlapping, their separation into individual components
provides an opportunity to emphasize several points.
3.2 Capacity Design of Connections and Connectors
In order to insure that primary structural elements such as beams and columns reach their full
capacity, the connections between such elements may be intentionally proportioned to exceed the
strength of the individual members regardless of the calculated demands. Also, there are critical
connector elements, which collect and transmit lateral forces to a lateral resistance system or which tie
individual systems such as coupled shear walls that are proportioned in this manner, i.e. to develop the
full strength of the member at the point of connection regardless of the forces calculated from the design
loads.
It should be recognized that this design philosophy can result in unanticipated drastic changes in
the size and weight of certain members unless the structural configuration is altered accordingly. In [8],
it is shown that the size of the floor beam in a steel frame with conventional V or Chevron bracing may
increase to three times the weight required by the vertical design forces because a large vertical seismic
load is introduced at the mid-span by the difference in the tensile yield strength of one member and 30%
of the compressive strength of the other intersecting member, as required by ANSI/ASCE 341-10
Seismic Provisions for Structural Steel Buildings. While this design meets the intent of the provisions,
it does not reflect creative engineering. The proposed alteration was to use alternating Chevron and V
bracing intersecting at the centerline of the floor beam, which essentially creates a two story X brace
that provides the counter resistance for the unbalanced forces arising from the capacity design. There
are undoubtedly other cases where the traditional configuration of a system may need to be altered to
avoid a drastic size change in a key element beyond the requirements of the loading demands.
3.3 Ductile Detailing
In order to provide a ductile load path prior to any possible failure regardless of the cause, the
connections and the connectors mentioned in Section 3.2 and the structural elements themselves should
be detailed appropriately. In many cases the intention is to direct such a failure from the connection
region to the members themselves, such as forcing the plastic hinge in a steel beam away from the
beam-to-column connection. Obvious enhancements of ductile detailing include continuity plates and
doubler plates in steel connections and closed beam stirrups and transverse joint steel in concrete
connections.
51
KEYNOTE PAPERS
However, the obvious locations requiring enhanced detailing may not be so obvious in wall and
tube-type structures. Nonlinear pushover analysis, described in Section 4, revealed that the large
opening near the base of the chimney shown in Fig. 1, purportedly the location of the initiation of the
failure, was insufficiently reinforced to resist the demand of the shear stresses in the immediate vicinity
as shown in Fig. 5 Similarly the loss of the roofs of the silos shown in Fig. 3(c) indicates insufficient
reinforcement between the roof and the wall to withstand a blast loading.
Fig. 5: Cracks around Opening: Loading 90° to axis (Huang, et.al, 2004)
3.4 Continuity
Load-path continuity that conforms to the relative stiffness and overall strength of the structural
elements in the lateral force-resisting system along with specific strengthening of individual
components and connections, such as those noted in the preceding sections, is essential for resisting the
effects of extreme loading.
Design procedures for structural steel members generally include the provision of adequate lateral
and local bracing to prevent premature failure or discounting the capacity of the member when
sufficient bracing to develop the full strength is not feasible. While the strength demands on bracing
elements may not be large for conventional serviceability behavior, the demands from an extreme
loading case could change this drastically.
For reinforced concrete members, which are stockier than comparable steel members, bracing
demands are not as severe. However, it is advisable to enhance continuity by increasing development
length, splice lengths and hook requirements along with the additions noted in Section 3.3.
3.5 Transfer Mechanisms
In order for structures to survive extreme loading at force levels beyond the standard design
requirements, all available resistance mechanisms may be required. Most structural design procedures
for lateral loads are based explicitly or implicitly on two-dimensional conceptual models. The survival
of some structures that were subjected to forces far beyond the conventional design values may be
attributable to three-dimensional actions mobilized by the interchange and sharing of loads between
designated resistance systems. Important elements in buildings for affecting this transfer are floor and
roof diaphragms, which may not receive the attention of other primary components of the lateral
force-resisting system.
Fig. 6 shows a reinforced concrete building that collapsed in the 2010 Haiti earthquake. The
catenary action of the floor slab, which also serves as a diaphragm, is evident at the first and second
floors. Despite the collapse of the exterior column between the ground and first floors, the upper floors
did not collapse due to this secondary action and perhaps some redistribution through the floor slab to
the interior framing lines. There is apparently some support to the first floor provided by the remaining
wall. Unfortunately the author did not have access to this structure or to the drawings, but it is apparent
that that a three-dimensional analysis would be required to fully explain the response.
52
KEYNOTE PAPERS
Fig. 6: Damaged Reinforced Concrete Building
4. Nonlinear Analysis Procedures
4.1 Methods for Time-Varying Loading
For wind loading, even of extreme intensity such as a tornado, quasi static analysis seems to be the
most popular approach. Nonlinear time history analysis (NLRHA) is widely regarded as the gold
standard for estimating the response of a structure to extreme earthquake and blast loading. Since
NLRHA is time consuming and often costly, it has been most popular in research and for special and
unique structures and less so for more routine design. However it is being more widely used as
computational resources become more affordable and if a well-established suite of time histories is
available. Representative time histories for blast and earthquake loading are shown in Fig. 7.
Fig. 7: Time Histories for Blast and Earthquake Loadings (Sammarco et.al, 2014)
53
KEYNOTE PAPERS
NLRHA investigations are also performed when the adequacy of an existing structure for a loading
condition not considered in the original design is of interest, or if the structure has been overloaded in
some way with the potential for resulting damage or failure.
It is appealing to seek simpler dynamic methods for the evaluation of the response of structures to
extreme lateral loads. Nonlinear static procedures (NSP), especially pushover analysis, have become
popular in structural engineering practice. The analysis consists of two stages termed Demand and
Capacity, which are ideally compared in a common format.
The capacity curve is initiated by applying a monotonically increasing lateral load pattern to the
structural model in stages and carrying out a static analysis at each stage using the updated material and
geometrical data to determine the displacement of a reference point, the performance point. Since the
structure generally becomes more flexible due to material softening and geometric distortion, it is
eventually pushed over to failure at some multiple of the reference initial loading.
The demand curve may be conveniently generated using the capacity spectrum method (CSM).
The CSM was developed for seismic loading [9] and also has been applied in limited blast loading
investigations [10,11]. Referring to Fig. 7, the earthquake loading has obvious periodicity while the
blast loading shows a nearly instantaneous peak overpressure followed by a slower decay to a negative
pressure as described in Section 2.3. The corresponding time histories may be transformed into
traditional response spectra (RS) plotted as spectral acceleration (A) vs. period (T). These spectra are
converted from the acceleration-period (T) domain to the acceleration-displacement domain (ADRS)
based on the simple harmonic motion relation for a single-degree-of freedom (A = ω²D) or by an energy
equivalence [10], and are augmented by modal corrections for multi-degree-of-freedom systems [11].
With the capacity also plotted in the ADRS domain, an intersection of the curves at the performance
point indicates that the system can resist the corresponding acceleration and displacement as shown in
the idealized diagram Figure 8[10]. No intersection would indicate a possible failure.
Fig. 8: Idealized ADRS Diagram (Lam et.al, 2004)
For the case shown in Fig. 8, the demand curves were computed for a blast scenario on a wall panel
3m high by 1m wide with a charge weight based on TNT equivalence of W= 125kg. The standoff
distance is R= 10m. Using a SDF substitute structure model, the acceleration time history is
transformed into a traditional acceleration vs. period Response Spectrum (RS) and then represented in
the ADRS domain as noted in the previous paragraph. The simplified curve also shown is said to be a
conservative representation of the same blast scenario as is the 5% damping level [10].
The capacity curve shown was obtained from a pushover analysis of a wall with a total mass of
2.26 tonnes, effective mass (parameter in SDF model mentioned in preceding paragraph) of 1.7 tonnes
and effective natural period of T=0.05sec. The effective stiffness is about 27 MN/m [10]. The force on
the wall was found by applying pressure on one side of the wall in small increments (either by use of an
air bag or by numerical simulations), and calculating the total force applied in each increment. The total
force in each increment is then correlated with the effective displacement of the deflecting wall to
obtain the capacity curve [12].
It is pointed out in [7] that the material properties for a rapidly applied loading may be appreciably
altered from those commonly assumed for static and quasi static loadings, as shown in Fig. 9 for the
strain rate of concrete, and this would be reflected in the construction of an appropriate capacity curve.
However, this refinement appears to be uncommon for earthquake loading.
54
KEYNOTE PAPERS
Fig. 9: Strain Rate Effects for Concrete (Sammarco et.al, 2014)
The CSM procedure was applied to analyze the Tupras stacks described in Section 2.2 and shown
in Figure 1. Of course, the demand on a particular structure could only be described accurately if there
were instruments at the site during an earthquake event - a rare occurrence. While there was a ground
motion record in the vicinity of the chimneys, the resulting spectrum YPTy is only an indicator of the
motion at the site due to local soil conditions and other factors, but it is likely better than most data
available for analysis at similar industrial sites that were impacted by the strong ground motion. Two
smoothed spectra of the type often used for design are also shown in Fig. 10, YTPy Mean and YTPy
Mean plus one standard deviation. Also plotted is a spectrum corresponding to the UBC code for
comparison to what might be used for a design in a high seismic region.
The results shown in Fig. 10 are for an excitation applied perpendicular to the axis of the opening
[1]. The corresponding capacity curve is constructed based on a shape that considers the first three
mode shape (SRSS) and shows the decrease in stiffness as the pushover progresses, with the phases
indicating moderate to severe cracking. Note that the capacity curve does not reach the demand of any
of the smoothed spectra, confirming a possible failure. Phase 3 on the capacity curve corresponds to
Fig. 5 and is consistent with the observed shear failure.
1.60
1.40
T = 1.0 s
1.20
Spectral Acceleration (g)
1.00
T = 2.0 s
0.80
0.60
0.40
0.20
0.00
0 20 40 60 80
Pha.1 Pha.2 Pha.3
Spectral Displacement (cm)
YPTy YPTy Mean YPTy Mean+1 STDV UBC-97 ModeSRSS Withhole 90 Deg.
Fig. 10: Capacity Spectrum Analysis for Tupras Chimney Loaded 90° to opening (Huang, et.al, 2004)
55
KEYNOTE PAPERS
5. Conclusions
 The three diverse examples selected demonstrate that structures designed and built for code
or practice- specified requirements may fail under unanticipated extreme loading. These
requirements may implicitly ignore loading conditions that are judged to have an extremely
low probability of occurring and may not fully address local stress concentrations. In each
example however, the failure can be attributed to a loading condition or stress concentration,
perhaps beyond these requirements but well within the realm of possibility.
 The strength and toughness principles presented transcend the particular type of structure and
loading. It is emphasized that these principles apply not only to the primary structural system
but to secondary systems, cladding, windows, and mechanical and architectural components.
A structure need not collapse to functionally fail due to serviceability disruptions.
 As a structure is loaded beyond the service load range by an extreme load, the response
changes from linear to nonlinear until a failure occurs or the structure survives, likely with
residual damage. Modern analytical techniques enable the structural behavior to be assessed
for a variety of loading conditions, some of which may not have been included in the original
design requirements. In turn, significant potential structural weaknesses may be identified
and reduced or eliminated, regardless if the loading conditions that would excite the
weakness were included in the design criteria. This type of assessment can guide efforts to
improve the performance of the structure under specified or unanticipated loading
conditions.
Acknowledgments
The careful reading and suggestions of Dr. Nathan Gould, P.E., S.E. are greatly appreciated.
References
[1] Huang, Wei, Gould, Phillip L., Martinez, Raul, and Johnson, Gayle S., Nonlinear analysis of a
Collapsed Reinforced Concrete Chimney, Earthquake Engineering & Structural Dynamics,
Vol. 33: pp.485-498, 2004.
[2] Primer for Design of Commercial Buildings to Mitigate Terrorist Attacks, FEMA 427,
December 2003.
[3] Corley, W.G., Private Communication
[4] DeBruce Grain Elevator Explosion Report, https://www.osha.gov/SLTC/grainhandling/geeit/.
[5] Kuligowski, E.D., Lombardo, F.T., Phan, L.T. and Levitan, M.L., Final Report, National
Institute of Standards and Technology (NIST)Technical Investigation of the May 22, 2011
Tornado in Joplin, Missouri, March2014
http://nvlpubs.nist.gov/nistpubs/NCSTAR/NIST.NCSTAR.3.pdf.
[6] http://www.ssestructural.net/press/
[7] Sammarco, E.L., Jones, C.A., Williamson, E.B. and Sprague, H.O.. Design for Blast and
Seismic, STRUCTURE magazine, pp. 10-13, March, 2014.
[8] Johnson, J.G. How Big is that Beam? , STRUCTURE magazine ,pp 44-45, November, 2014.
[9] ATC-40, Seismic Analysis and Retrofit of Concrete Buildings, Vol.1, Applied Technology
Council, Redwood City, CA, November, 1996.
[10] Lam, N., Mendis, P. and Ngo, T. Response Spectrum for Blast loading, Electronic Journal of
Structural Engineering, 4, 2004.
[11] Walker, S., Corr V., Tam, B, and Shearer, M, Response Spectra for Explosion Loading and
Response , OTC 17242, Houston TX, May ,2005.
[12] Lam, N., Private Communication
56
KEYNOTE PAPERS

0940 - 1020 hours
Understanding the Boundary Condition when Designing Protective

José L. Torero
School of Civil Engineering, The University of Queensland, Australia

(j.torero@uq.edu.au)
Protective structures are designed explicitly to fulfil a function. The function is in many cases an
extreme event, therefore, the explicit design has to properly and precisely account for the nature of
the solicitation imposed by the extreme event. Extreme events such as explosions or earthquakes are
reduced to design criteria on the basis of either empirical or historical data. To determine the design
criteria, the physical data has to be translated into physical variables (amplitudes, pressures,
frequencies, etc.) that are then imposed to the protective structure. Performance is then quantified on
the basis of the structure’s capability to perform its function. If the input variables can be classified,
only then it is possible to translate performance into code requirements that implicitly guarantee that
the structure performs its function. When addressing fire, performance has been traditionally
determined by imposing standardized requirements that only bear a weak relationship with possible
events. The performance of a protective structure is thus defined as a Fire Resistance period. This
paper addresses the concept of Fire Resistance and its relevance to the design of protective structures.
Keywords: Fire; Fire resistance; Protective design; Explicit performance
1. Introduction
Whatever is the function of a protective structure, its capability to perform its function is defined
by a design process that needs to contemplate the different solicitations that these structure may have.
In many cases this requires the understanding of single and multiple solicitations. Critical
infrastructure is generally designed to be able to withstand combined hazards, therefore, protective
structures are also generally designed to perform their function when affected by combined hazards.
While protective structures can be designed for fire, it is often necessary to introduce fire as part of a
multiplicity of hazards. This is the case of mine explosions and protective seals, where fires tend to
follow explosions, or that of the structures that protect the core of nuclear reactors that could be
subject to terrorist attacks or earthquakes followed by fires. Protection to fire, in the case of combined
hazards, makes the assumption that the structure’s capability to withstand the fire remains intact. This
assumption has the potential of not being realistic, nevertheless, because of the nature of the design
process for fire, it is unavoidable. When designing for fire performance is not explicitly calculated
but is calculated on the basis of a presumption that “Fire Resistance” represents a worst case fire
scenario condition that if applied to a single element of a structure will deliver a solicitation that
cannot be exceeded by any realistic fire. The solicitation and its manifestation on the behavior of the
structure is defined in purely thermal terms and does not require any quantification of structural
performance. Some element of structural performance might be part of the standard “Fire Resistance”
testing procedures when addressing a single element (deflections under load, penetrating cracks, etc.)
nevertheless how the element will perform under fire as part of a structural system is never addressed.
Once the requirement of “Fire Resistance” is met, then it is assumed that all serviceability
requirements for the structure will be met independent of any other solicitation or the integral nature
of the structural system.
57
KEYNOTE PAPERS
A very good example of a protective structure where the inconsistency between design for other
hazards and fire is well known is that of coal mine seals. Coal mines are environments where fire
hazards coexist with explosion hazards. These can manifest themselves in different manners. An
explosion can lead to a fire or a fire can lead to an explosion. Seals should be designed to minimize
the consequences of either event or combined events. The main purpose of seals is to isolate
worked-out areas and active fire zones [1] but have been designed on the basis of an analysis that
fundamentally incorporates explosion hazards and only summarily addresses the potential of a fire as
a function of mild requirements of “Fire Resistance” [2]. When it comes to explosion design a series
of large scale experiments have shown that the magnitude of the pressure rise in an explosion, a
Deflagration to Detonation Transition (DDT) or Detonation can vary within at least three orders of
magnitude [3, 4, 5]. This fact has been known for over thirty years thus regulations have accepted that
seals are to be designed to contain flammable regions within the sealed zones but are not to be
expected to withstand all possible pressure rises associated to potential explosions, DDT or
detonations. The main parameters controlling this variation are well-known and are build-up length,
stoichiometry and homogeneity of the flammable mixtures, and induced turbulence. All these
parameters are coupled and a function of many other secondary parameters, thus isolating the root
cause for a specific overpressure is not possible. Therefore, Mitchell’s recommendation [2] to design
seals to resist an explosion with an overpressure of 20 psi is intended to ensure containment of the
flammable mixture and to provide protection against pressure increases for a large range of possible
explosions, but not to guarantee that a seal will withstand the pressure increases associated with all
possible explosions, DDT or detonations.
Until the Sago explosion of January 2, 2006 it was accepted that 20 psi seals were adequate and
sufficient for their intended purpose, thus the focus of research was to establish better ways to
construct and assess performance of those seals to the defined value of 20 psi. The National Institute
for Occupational Safety and Health (NIOSH) issued a report in which they were working towards
alternative methods for testing the 20 psi seals, rather than working towards changes in the
regulations or the development of novel engineering methods to determine more appropriate pressure
increases [6]. An earlier research effort, also by NIOSH, was also geared towards creating small scale
tests that could replace the expensive large-scale tests [7]. Even after the Sago explosion of January 2,
2006, Sapko et al presented another report on alternative testing methods [8]. Still, the focus is not
prevention of the incident or on the need for stricter regulation requirements for seal resistance, but
rather on creating test methods to see if seals could withstand the 20 psi as suggested by Mitchell in
1971 [2]. While criticisms towards the 20 psi requirement and the tests used to establish how this
pressure is to be applied are many, it is clear that the approach towards explosions is one that
incorporates a quantitative definition of the load and an adequate testing procedure to define the way
this load is applied. Nevertheless, while the functional requirement of coal mine seals is to address the
combined fire and explosion hazard, the fire aspects of existing regulations and procedures do not
include any quantitative means of defining the solicitations to be imposed by a fire of the combination
of both hazards. This paper will help define the validity of the “Fire Resistance” framework when
addressing this particular issue so evident in coal mine seal design.
The literature addressing the structural design of protective structures in the absence of thermal
loads is extensive and so is the literature related to the behavior of structures (including protective
structures) under thermal loads. This paper does not intend to provide a review of these subjects
instead it will address a different issue and this is the weaknesses of current performance assessment
methods when it comes to the determination of the performance of protective structures when subject
to fires and combined hazards that include fires. The focus will be on concrete structures because of
their great relevance to protective structures but also because the potential shortcomings of current
design methods might have more dramatic manifestations.
2. Assessing Thermal Performance
2.1 Fire Dynamics
At the core of a fire there is a flame or a reaction front that is effectively a combustion process,
and thus is governed by the mechanisms and variables controlling combustion [9]. The interaction
between the fire and the environment determines the behavior of the flame and nature of the
58
KEYNOTE PAPERS
combustion processes. This is commonly referred to as Fire Dynamics. An extensive introduction to

the topic is provided by Drysdale[10].
As indicated by Drysdale[10], Fire Dynamics involves a compendium of different sub-processes
that start with the initiation of a fire and end with its extinction. The onset of the combustion process,
i.e. ignition, in a fire is a complex process that implies not only the initiation of an exothermic
reaction but also a degradation process that provides the fuel feeding the fire. In a fire it is common to
have different materials involved and given the nature of the fire growth many could be involved
simultaneously but others sequentially. The sequence of ignitions of items in an enclosure will affect
the nature of the combustion processes. Thus, ignition mechanisms set the dynamics of the fire and
also are affected by the fire itself, creating a feedback loop. The process of ignition is reviewed in
detail by Torero [11].
Once a material is ignited, the flame propagates over the condensed fuels by transferring
sufficient heat to the fuel until a subsequent ignition occurs. This process is commonly referred to as
flame spread and is described in detail by Fernandez-Pello [12]. Flame spread defines the surface area
of flammable material that is delivering gaseous fuel to the combustion process. The quantity of fuel
produced per unit area is the mass burning rate. The mass burning rate multiplied by the surface area
determines the total amount of fuel produced. If the total amount of fuel produced is multiplied by
the effective heat of combustion (energy produced by combustion per unit mass of fuel burnt), it
yields the heat release rate. The heat release rate is generally considered the single most important
variable to describe fire intensity [13]. Given the nature of the environment, the oxygen supply might
not be enough to consume all the fuel, thus in many cases combustion is incomplete and therefore the
heat of combustion is not a material property but a function of the interactions between the
environment and the fire. In these cases it is usually deemed appropriate to calculate the heat release
rate as the energy produced per unit mass of oxygen consumed multiplied by the available oxygen
supply.
If the fire is within a compartment, smoke will accumulate in the upper regions of the
compartment. Hot smoke will radiate and/or convect heat towards all surfaces. If the surfaces are
flammable, remote ignition of different materials can occur. If remote ignition occurs in the lower
(i.e. cold) layer then the fire tends to suddenly fill the entire compartment. This transition is generally
named flashover. Before flashover, the lower layer tends to have enough oxygen to burn the
pyrolyzing fuel and the heat release rate is determined by the quantity of fuel generated. This period is
termed pre-flashover, fire growth or fuel limited fire. After flash over, fuel production tends to exceed
the capability of air to enter the compartment, the compartment becomes oxygen starved and the heat
release rate is determined by the supply of oxygen (i.e. doors, windows, etc.). This period is termed
post-flash over, fully developed fire or oxygen limited fire. The process of fire growth and the
definition of the different variables affecting it is provided by Drysdale[10].
For small compartments (approximately 4 m x 4m x 4m) a characteristic time to flashover is of
the order of 4-6 minutes while the post flashover period can reach tenths of minutes depending on the
compartment size and fuel available. Structures tend to have high thermal inertia, thus the
temperature increase to levels where ignition of the structural elements as well as the loss of
mechanical properties takes also tenths of minutes. Thus for purposes of structural assessment, fires
tend to be considered as post flashover fires [14]. The temperature of the compartment as well as the
burning rate can be established simply as a function of the available ventilation, this process can
follow different levels of complexity and these are all reviewed by Drysdale[10].
A fire can end when it is extinguished or when oxygen or fuel supplies are depleted (oxygen
starvation and burnout respectably). In all cases, extinction of the combustion process is brought by
interactions of fuel and oxygen supply and the energy balance that permits the combustion reaction to
remain self-sustained [15]. Suppression agents affect a fire by reducing fuel and oxygen supply or by
removing heat. At each stages of fire growth, it is more or less feasible to affect these three variables.
Thus the effectiveness of a suppression system is dictated by its capability to affect the targeted
variables at the moment of deployment.
A common way to describe the evolution of a fire is by mean of design temperature vs. Time
curves. There are many variants of these, from a purely standardized version [16] used to conduct
furnace tests to others more representative of real fire conditions [17]. A commonly used expression
is that proposed by Lie [18] and simplified here for small fuel loads
59
KEYNOTE PAPERS
.
.
250 10 exp . 3 1 exp 0.6 1 exp 3 4 1 exp 12 (1)
Where Tg is the temperature of the gas phase inside the compartment and t is time in hours. F is
the ventilation factor, which is a commonly used combination of the opening surface area (AW), the
height of the opening (H ) and the total area of the compartment (AT) excluding floor and opening
(F=AWH1/2/AT) [10]. Fig. 1 describes a typical scenario where all critical conditions have been
marked. The maximum heat flux has been calculated assuming that the solid surface remains at
ambient during heating and at Tg during cooling. The gas phase temperature is assumed to be 20oC
during cooling. The total heat transfer coefficient, hT= 45 W/m2, is a value commonly used to account
for convection and radiation by fire engineers. While clearly, the surface and gas phase temperatures
will vary during heating and cooling, this value of the heat flux is indicative of the conditions that a
structure will experience in a fire.
Fig. 1: Typical temperature and heat flux evolution for a small compartment fire. From ignition to flashover there is a
period of approximately 5 minutes where temperatures and heat fluxes are negligible. This is followed by a longer period
where temperatures and heat fluxes increase. The cooling period starts with burn-out (total fuel consumption). The plots
were obtained using a 4 m x 4 m x 4 m compartment with a single opening (2 m x 3 m) a fuel load of 60 kg/m2 and a total
heat transfer coefficient of 45 W/m2. The temperature vs. Time curve was obtained using the expression by Lie [1], Eq. (1).
In summary, structures exposed to fire will see temperatures close to ambient and negligible heat
fluxes during the pre-flashover period (4-6 min for a small compartment). Temperatures and heat
fluxes will increase after flashover reaching values of approximately 1200oC and 60 kW/m2
respectively. These exposures can last for periods in excess of an hour, depending on the fuel load,
compartment geometry and ventilation. Finally, fuel will burn-out (in Fig. 1, 60 kg/m2) and the
compartment will cool down. In the cooling period the heat flux will be negative denoting heat losses
from the structure surface to the colder air. Thus the boundary condition for the concrete surface is
defined by means of Eq. (2)
" (2)
As can be seen in Fig. 1, 60 / " 60 / therefore the gradients of

temperature within the structural elements can vary significantly and in a manner that does not have
to resemble the temperature evolution. The importance of the gradients and the resemblance of their
evolution to that of the temperature are a function of the thermal properties of the material. Therefore,
when trying to understand the explicit behaviour of structures in fire it is essential to discuss its
evolution not only within the context of the characteristic conditions and time scales involved in a fire
but also as a function of its thermal properties. While the analysis presented above is simplistic in
60
KEYNOTE PAPERS
nature, and the values presented are only rough estimates, it does provide a clear image of the
conditions that any construction material will see in the event of a fire.
2.2 Furnace or Fire?
When analysing the heat transfer from the fire to a structural element the problem needs to be
formulated in terms of heat fluxes. While temperatures result from the solution to the energy
conservation equation, all quantities to be balanced are energies [19].
Heat is transferred from gases to surfaces via radiation and convection resulting in a total heat
flux, " , where
" " " (3)
And " is the heat transfer via radiation and " is the heat transferred via convection. For
simplicity the problem will be assumed as one dimensional, then the boundary condition for a solid
element (structural element) becomes
" (4)
Which is a generic version of Eq. 2 and where the thermal conductivity (ki) is a property of the solid
and the gradient of temperature is taken at the surface. In other words all the heat arriving is
conducted into the solid. If there are multiple layers then at each interface the following boundary
condition should apply:
(5)
Where the gradients correspond to each side of the interface and the sub-index “s” is a generic way to
represent the next layer of solid. Once the boundary conditions are defined, the energy equation can
be solved for each material involved. In the case where two layers of solid are involved (“i” and “s”),
then the energy equations take the form
(6)
(7)
The solution to the energy equation will result in the temperature evolution of the material in
space and time. Eqs.6 and 7 could be repeated for as many layers as necessary.
If the geometry is complex, then the problem needs to be resolved in all dimensions. If the
properties vary with temperature then, as the temperature increases, these properties need to evolve
with the local temperature. Variable properties thus require a numerical solution. If a simple
analytical solution is to be obtained then adequate global properties need to be defined. It is
important to note that whatever the solution methodology adopted, the temperature of the structure is
the result of the resolution of Eqs.6 and 7 using boundary conditions such as those presented in Eqs.4
and 5.
To obtain the numerical solution it is necessary to input material properties for the different
layers (“i” and “s”). The material properties required are all a function of temperature and are as
follows:
ρi, Cpi, ki
ρs, Cps, ks
61
KEYNOTE PAPERS
For some materials such as steel the properties of are well known and thus very little difference
can be found between the literature [20]. For other materials such as concrete, wood or different
thermal insulations the scatter is much greater [14]. The uncertainty is associated to the presence and
migration of water, degradation, crack formation, etc.
Furnace data is generally used as a substitute for the uncertainties associated to property
definition. In many cases global properties are extracted by fitting temperatures to the furnace data.
These properties are then extrapolated and many times used in equations such as Eqs. 4 to 7 for
performance assessment. Nevertheless, this practise has also its unique complexities. First of all the
model needs to include all the physical variables necessary, so if physical processes such as the
degradation or water are not included in the model, the properties used from the furnace calibration
become hybrids that include these physical parameters. Introducing physical phenomena into
constants inevitably narrows the range of application, thus most of these calibrated properties can
only be used to re-evaluate furnace data. Extrapolation to drastically different scenarios such as a fire
becomes doubtful.
An important aspect many times overlooked is the need to make sure that the boundary
conditions are properly represented. The heat exchange between a furnace and a sample is extremely
complex and many times simplifications relevant to furnaces are not valid for fires. It is essential to
understand all those simplifications. The differences between furnace behaviour and fire behaviour
are all manifested in the boundary condition associated with Eq. 2.
A common misunderstanding is to attempt representing the evolution of the temperature of a
material by a single temperature as represented in Fig. 2. Fig. 2 shows the evolution of a global
temperature for unprotected steel for different fires. While plots of this nature serve to compare the
evolution of the steel temperature they hide numerous assumptions that while relevant to steel, they
are not relevant to other materials, in particular concrete.
Establishing the nature of temperature gradients within a solid allows to validate the assumption
of a single temperature representation and is an essential component of the analysis of a material in a
fire. The nature of the temperature gradients is defined by the Biot number
(8)
Fig. 2: Temperature evolution of the gas phase of a compartment fire and a small cross section unprotected steel beam.
Standard temperature time curve per ISO-83 [16], DFT stands for Dalmarnock Fire Test [21].
Fig. 3 provides a very simple schematic of the role of the Biot number in one dimensional heat
transfer and the potential simplifications. If the Biot number is close to one (case (b) in Fig. 3)
temperature gradients in the gas and solid phases are large and therefore Eqs 6 and 7 will need to be
fully resolved. No simplifications are possible. In case (c) (Fig. 3) the Biot number is much greater
62
KEYNOTE PAPERS
than one therefore the temperature differences in the gas phase are much smaller than those in the
solid phase and it can be assumed that surface and gas temperatures are almost the same. This
simplification is very important when modelling furnace tests because it enables to ignore the
complex boundary condition imposed by the furnace and simply imposed the monitored gas
temperature at the surface of the solid. Finally, if the Biot number is much smaller than one (case (a)
in Fig. 3) then the temperature differences in the solid phase are much smaller than those in the gas
phase, therefore temperature gradients in the solid phase can be ignored and a single temperature can
be assumed for the solid. Heat conduction can be approximated by the boundary condition and Eqs. 6
and 7 lead to a single temperature solution like the ones in Fig. 2. The representation of a structural
element by means of a single temperature is therefore only valid if Bi <<1. This simplification is
called a “lumped capacitance solution” and while it does not resolve spatial temperatures
distributions it still requires an adequate definition of the heat transfer between the furnace and the
sample. An important observation is that for materials with Biot numbers much smaller than one, the
energy is rapidly diffused through the integrity of the material, so if the density was to be high, then
the lumped solid will lag significantly the gas phase temperature (Fig. 2). Heat transfer is therefore
dominated by the temperature difference between the solid and the gas phase and errors in the
definition of the heat transfer coefficient loose importance.
Given the importance of the Biot number in the characteristics of the temperature gradients, it is
important to estimate the thickness of a material that leads to a Bi=1. Samples much thicker will
allow approximating the surface temperature to that of the gas phase. Samples much thinner will
allow to “lump” the solid phase into a single temperature.
(a) (b) (c)
TH TH TH
TS ≈ TH
d d
TS
TS ≈ TC TC TC
TC
d
≪1→ ≫1→
1
Fig. 3: Schematic of the typical temperature distributions for different extreme values of the Biot number.
Table 1 shows typical thermal properties for different construction materials and the
characteristic thickness (d) that will result in a Biot number of unity. As can be seen for high thermal
conductivity materials like aluminium or steel, sections a few millimetres thick can be lumped
without any major error. In a similar manner very low thermal conductivity materials like
plasterboard or EPS will allow to assume that the surface temperature of the solid is that of the gas
phase. In contrast, concrete has a Biot of unity for a thickness of 50 mm which in between typical
concrete cover thicknesses and the overall thickness of the sample. So whether the concrete is used as
cover for the reinforcement or analysed as the load bearing material, the full resolution of an equation
similar to Eq. 6 is necessary. Furthermore, the boundary condition cannot be simplified because the
thermal gradients are fully defined by " as per Eq. 4.
63
KEYNOTE PAPERS
Table 1: Typical thermal properties for different construction materials

Material Density Thermal Specific Thermal “d” for
(ρ,kg/m3) Conductivity Heat Diffusivit Bi=1 (m)
(k,W/mK) (Cp,J/kgK) y (α,m2/s)
Aluminium 2,400 237 900 1.10E-04 5.30
Steel 7,800 40 466 1.10E-05 0.90
Concrete 2,000 2.5 880 1.42E-06 0.05
Plasterboard 800 0.17 1,100 1.93E-07 0.004
Expanded polystyrene (EPS) 20 0.003 1,300 1.15E-07 0.0001
The resolution of the thermal fields in a structural element defines differential thermal expansion
and thermal-gradient induced stresses. As demonstrated above, these thermal fields require the full
resolution of the heat transfer equation with the appropriate boundary condition ( " ). The thermal
boundary condition provided by a furnace is therefore relevant only if it is consistent to that of a fire.
3. The Structure
Once the thermal boundary condition (heat fluxes and thermal gradients in the structural
elements) any form of structural analysis (analytical methods, Finite Element Analysis, etc.) can be
performed. The structural analysis will establish the true performance of the structure and enable to
define if the structural design needs to be changed or protected more or less. It is important to note
that the protection then becomes a variable of the analysis. Fire protection can be enhanced to protect
unusually vulnerable structures or reduced for those which present a robust behaviour in fire.
When it comes to the design of concrete structures, current design procedures are to meet certain
minimum cover and size limits according to a number of deemed-to-comply rules [16] to achieve the
required fire-resistance period. Such an approach does not require any temperature-dependent
calculations. As described above, furnace thermal loadings or test results will not deliver the
performance of a concrete structure in fire, even if complex heat transfer calculations or finite
element structural models are developed. The performance of concrete in fire is critically linked to
the generation of thermal stresses and therefore without a detailed knowledge of temperature
gradients these will never be realistic. Even if the fire and heat transfer were to be incorporated, the
models used to calculate the performance of concrete structures require a series of parameters that are
extracted from experiments. Among these parameters are the temperature-dependent properties of
concrete and reinforcement. Due to the high thermal conductivity of steel, the reinforcement
properties are easily determined and thus readily available. Relevant properties of concrete, however,
are not well addressed due mainly to the considerable problems to be overcome in their experimental
determination. Typical properties among these include concrete strength, Young’s modulus of
elasticity, strain at peak stress, creep and stress-strain relationships. Currently, these properties are
inferred from available experimental data generated through furnace/oven heating [22]. When
heating with an oven/furnace, the heat-flux absorbed by the sample is the result of a complex heat
exchange between the furnace and the sample. Different heating rates will result indifferent heating
histories for the samples; nevertheless these histories will be complex in nature and only defined if
the time evolution of the in-depth temperatures of the sample is fully characterized. Characterising
in-depth temperatures implies intrusive (i.e. thermocouple) measurements that if introduced in
sufficient quantity will affect the mechanical properties of the sample, therefore gradients are never
characterized and tests are conducted at heating rates that minimize temperature gradients. As a
result, the derived correlations between the observed concrete properties and temperature are only
approximate.
Models that describe the high temperature performance of a structure and have as inputs
parameters extracted from the tests will have the following limitations:
 Mass transfer processes affected by heat will be very different because the very slow heating
rates will not only allow dissipation of heat through the sample but also slow dissipation of
water vapour with minimal pore pressure increase.
64
KEYNOTE PAPERS
 The temperature gradients while small remain undefined.

 None of the components of the model linked with temperature gradients will be properly
addressed.
 None of the couplings between different processes linked to temperature gradients (i.e.
moisture transport, vapour pressure, thermal gradient induced stresses, etc.) will be
addressed.
4. Conclusions
A simple framework for the introduction of structural fire behavior into the design of structural
systems is presented. This framework resorts to an explicit representation of the thermal boundary
condition provided by a fire as well as a comprehensive heat transfer analysis. Simplifications
associated to furnace testing are shown to provide an unrealistic representation of the thermal
conditions in a fire and prescribed fire proofing have been demonstrated to preclude comprehensive
optimization of structural systems. Constitutive properties of concrete are intimately linked to the
formation of thermal gradients and therefore, current values are at best approximate.
Protective structures are complex systems that require a precise and detailed representation of
their evolution as different solicitations are applied. In some areas such as explosions this is done in a
very careful way, and while questions still remain about the adequacy of the calculations and testing
procedures, all these are perfectly geared towards the explicit determination of performance. When it
comes to the explicit representation of performance in fire, current methods misrepresent the thermal
solicitation by the fire and tests that are intended to deliver material properties that serve as inputs for
models ignore some of the main parameters that determine the properties of the material. This is
particularly true for concrete structures.
Performance of protective structures has to be addressed in an explicit manner, this enables not
only to establish their reaction to a single hazard but their resilience when it comes to multiple
hazards. In the absence of an explicit performance assessment strategy for fire it is not possible to
determine the adequacy of the protection provided by a structure.
Acknowledgements
The author wishes to acknowledge the support of ARC through DP150102354 and
LE130100089 and the support of EPSRC through the “Real Fires for the Safe Design of Tall
Buildings” project.
References
[1] Sapko, M.J., Weiss, E.S. and Harteis, S.P. (2003) “Alternative Methods for Evaluating
Explosion Resistant Mine Ventilation Seals” Proceedings of the 30th International Conference
of Safety in Mines Research Institutes. Johannesburg, Republic of South Africa, 615-640.
(http://www.cdc.gov/niosh/mining/pubs/pdfs/amfee.pdf), NIOSHTIC-2 No. 20023644.
[2] Mitchell, D. (1971) “Explosion-Proof Bulkheads: Present Practices” USBM RI 7581.
[3] Cashdollar, K.L. and Sapko M.J. (2006) “Explosion Hazards of Coal Dust in the Presence of
Methane” in Handbook for Methane Control in Mining, Kissell, F.N. (editor), 147-150.
(http://www.cdc.gov/niosh/mining/pubs/pdfs/2006-127.pdf), NIOSHTIC-2 No. 20030254.
[4] Smith, A.C. and Sapko, M.J. (2005) “Detonation Wave Propagation in Underground Mine
Entries” Journal of the Mine Ventilation Society of South Africa, 20-25.
(http://www.cdc.gov/niosh/mining/pubs/pdfs/dwpiu.pdf), NIOSHTIC-2 No. 20030571.
[5] Zipf, R.K., Gamezo, V.N., Sapko, M.J., Marchewka, W.P., Mohamed, K.M., Oran, E.S., Kessler,
D.A., Weiss, E.S., Addis, J.D., Karnack, F.A., Sellers, D.D. (2010) “Methane-Air Detonation
Experiments at NIOSH Lake Lynn Laboratory” Proceedings of the Eighth International
Symposium on Hazards, Prevention, and Mitigation of Industrial Explosions, Yokohama,
Japan, 1-11.
65
KEYNOTE PAPERS
[6] Sapko, M.J., Weiss, E.S. and Harteis, S.P. (2005) “Methods for Evaluating Explosion Resistant
Ventilation Structures” Proceedings of the Eighth International Mine Ventilation Congress,
Brisbane, Queensland, Australia, 211-219.
(http://www.cdc.gov/niosh/mining/pubs/pdfs/mfeer.pdf), NIOSHTIC-2 No. 20028292.
[7] Sapko, M.J. and Weiss, E.S. (2001) “Evaluation of New Methods and Facilities to Test
Explosion-Resistant Seals” Proceedings of the 29th International Conference of Safety in
Mines Research Institutes, Katowice, Poland, 157-166.
[8] Sapko, M.J., Weiss, E.S., Harris, M.L., Man, C. and Harteis, S.P. (2010) “A Centennial of Mine
Explosion Prevention Research” SME Annual Meeting and Exhibit, Phoenix, Arizona, 1-13.
(http://www.cdc.gov/niosh/mining/pubs/pdfs/acome.pdf), NIOSHTIC-2 No. 20036611.
[9] Williams, F.A., “Combustion Theory,” Second Edition, The Benjamin/Cummings Publishing
Company, Inc., 1985.
[10] Drysdale, D. “An Introduction to Fire Dynamics,” John Wiley and Sons, 2nd Edition, 1998.
[11] Torero, J.L. “Flaming Ignition of Solids Fuels,” SFPE Handbook of Fire Protection Engineering,
4th Edition, Chapter 2-11, pp. 2-260-2-277, 2009.
[12] Fernandez-Pello, A.C., “The Solid Phase,” in “Combustion Fundamentals of Fire,” G. Cox
Editor, Academic press, 1995.
[13] Babrauskas, V. and Grayson, S.J. “Heat Release in Fires,” Editors, Elsevier Applied Science,
1992.
[14] Buchanan, A.H., “Structural Design for Fire Safety,” John Wiley and Sons, 2002.
[15] Torero, J.L. “Scaling-Up Fire,” Proceedings of the Combustion Institute, v. 34 (1), pp. 99-124,
2013.
[16] ISO, Fire Resistance Tests. Elements of Building Construction, ISO 834, International
Organization for Standardization: Geneva, 1975.
[17] Pettersson, O., Magnusson, S.E. & Thor, J., Fire Engineering Design of Steel Structures,
Publication 50, Swedish Institute of Steel Construction: Stockholm, 1976.
[18] Lie, T. T., “Characteristic temperature curves for various fire severities,” Fire Technology, 10,
315-326, 1974.
[19] Incropera, F. & DeWitt, D. (2002), “Fundamentals of Heat and Mass Transfer,” John Wiley &
Sons.
[20] Eurocode 1: Actions on structures – Part 1-2: General actions – Actions on structures exposed to
fire, EN 1991-1-2, CEN Brussels, Nov. 2002.
[21] Abecassis-Empis, C., Reszka, P., Steinhaus, T., Cowlard, A., Biteau, H., Welch, S., Rein, G.
and Torero, J.L. “Characterization of Dalmarnock fire test one,” Experimental Thermal and
Fluid Science, vol. 32, no. 7, pp. 1334 – 1343, 2008.
[22] Li, L.-Y. and J. Purkiss, Stress-strain constitutive equations of concrete material at elevated
temperatures, Fire Safety Journal, 2005. 40(7): p. 669-686.
66
KEYNOTE PAPERS
Wednesday, 21st May 2015

0900 - 0940 hours
Structures Subjected to Extreme Loading Conditions.

Do We Need New Knowledge?
Magnus Langseth
SIMLab – Centre for Research-based Innovation, The Norwegian University of Science and
Technology, Norway
(magnus.langseth@ntnu.no)
Structures can suffer from unintentional external loads and it is thus an increasing need to
protect critical infrastructure facilities and systems against terrorist acts, industrial accidents onshore
and offshore as well as from natural hazards such as floods, wind (storms, tornados, and hurricanes)
and rock fall on roads. Included here are also transportation of dangerous goods, road infrastructure
and car accidents. Addressing such challenges, may in fact, explain the reason for having structures
or systems which are able to withstand the defined threats, accidents and hazards in order to ensure
the probability of survival of people and other contents as well as reducing the economic losses and
the impact on the environment.
A rational approach for selecting appropriate protective measures is to carry out a risk
management analysis. The benefit of a comprehensive risk management approach is that the greatest
risks have been identified and addressed. Once the credible threats, accidents and hazards are
identified, a vulnerability assessment analysis has to be carried out. A key component of such an
assessment is properly defining the ratings for impact of loss and vulnerability, i.e. the attractiveness
of a facility as a target and the level of deterrence and/or defence provided by the existing
countermeasures. However, a proper and reliable vulnerability assessment requires a detailed
analysis of the response of the structure/facility subjected to the load from the defined incident. Here
a fundamental understanding of the behaviour of materials and structures subjected to high rate
loading is essential in order to evaluate potential risk to the facility from the threat, accident or
hazard.
Structures subjected to terrorist threats are often today designed based on existing design codes
for military facilities. However, careful attention must be devoted to use such regulations for civilian
facilities as essential knowledge is missing on how such facilities will behave for instance under such
extreme loading conditions. The reason for this is that civilian architecture differs from its military
counterpart by typically being lighter in construction, while at the same time more structurally
complex. In Norway the protection of critical infrastructure subjected to terrorist acts and sabotage is
guided by the Object Security Regulation [1]. The regulation states that each ministry shall appoint
objects to be protected and gives also general requirements with respect to protection. However, no
specific guidelines with respect to for instance structural protection are given as well as references to
actual design codes and regulations. The terrorist act in Norway the 22 July 2011 where a bomb
exploded in the government quarter in Oslo has put more focus than ever on this issue.
In the offshore and petroleum industry accidental actions are normally caused by abnormal
operations or technical failure such as explosions, impact from ships, dropped objects, fishing gear
impact, chance of intended pressure difference, helicopter crash etc. Such actions are accounted for if
the probability of an accident to occur is greater than 10-4. Design guidelines against such actions are
today covered by for instance the NORSOK standard [2], but such standards are limited in scope, i.e.
they cover a limited number of load and response scenarios and are thus today often supported by
advanced numerical simulations.
In the transportation industry, the design of vehicles and critical infrastructure along the roads
(included design for rock falls) are based on full scale testing which is time consuming and expensive
and also here numerical simulations have been introduced as an alternative to obtain safe design
solutions. Especially in the automotive industry, numerical simulations are an important tool which
now is replacing prototypes in order to reduce the product cycles, which in turn reduces time for
67
KEYNOTE PAPERS
design and development. For all events a realistic description of the load on the structure is crucial for
a proper design, even interaction effects have to be taken into account between the load and the
response.
Structural design for safety and physical security requires a sound background in fortification
science and technology and the basic knowledge is independent of the loading scenario [3]. Loading
environments associated with many relevant threats, accidents or hazards are extremely energetic and
their duration is measured in milliseconds. Thus, the structural response under short duration
dynamic effects could be significant different from much slower loading cases. Thus a designer must
explicitly address the effects related to such severe loading environments in addition to considering
the general principles used for structural design to resist conventional loads. Due to the large costs of
full scale testing, a strong emphasis has to be placed on the use of numerical simulations for load and
structural response calculations. Such numerical tools are independent of the problem at hand and can
thus be used for any design and risk assessments calculations, but also be used as input to more
simplified design tools and design codes. Even though great advances have been made in modelling,
the designer must still use knowledge about the physical mechanisms controlling the response. Here a
multi-scale testing and modelling approach to support the development of engineering models is
required. A structure or product can always be studied on different physical scales just like the
modelling scales. By using a top-down/bottom-up approach the main goal of the research will always
be the final structure or product. In some cases, microstructural modelling or even modelling on
atomic scale may be required to understand the underlying physical mechanisms of the observed
material response to loading, whereas for joints or components the behaviour may be sufficiently
well understood on the continuum scale. In all cases, research has to be designed to obtain modelling
frameworks on the material and structural levels that are suitable for industrial applications. Finally,
the designer must also know what simplifications can be made in the modelling and still retain
sufficient reliability and accuracy. At all levels of modelling, experimental validation of the
numerical models to be used is required before the models are accepted.
The present keynote lecture will give examples on research carried out at SIMLab – Centre for
research based innovation at NTNU during the period 2007-2014 [4]. The main focus will be placed
on the needs for new technology and why this knowledge is important. Examples will be taken from
materials and structures subjected to blast and fragment loading, research related to modelling of
aluminium in the automotive industry for crash applications and impact against submerged pipelines
in the oil and gas industry. The presentation will end with some guidelines on how industrial research
could be carried out in the future with focus on multi-scale testing, modelling and analysis for
industrial applications.
References
[1] Object Security Regulation. Ministry of Defence. LOV-1998-03-20-10.

[2] NORSOK Standard: Actions and Action effects. N-003, Rev. 1, February 1999.
[3] Krauthammer T.: Modern protective Structures. CRC Press, Taylor Francis Group.
ISBN 0-8247-2526-3, 2008
[4] SFI-SIMLab (www.ntnu.edu/simlab)
68
KEYNOTE PAPERS
Wednesday, 21st May 2015

0940 - 1020 hours
Design of Aircraft Structures for Protection of Bird Strike

Jun Liu, Yulong Li*, Zhongbin Tang and Xiancheng Yu
School of Aeronautics, Northwestern Polytechnical University, China

*
(liyulong@nwpu.edu.cn)
Bird strike accident has long been a significant threat to aircraft safety. In this study, design method
of structures under bird strike is introduced and summarized. When aircraft structures were initially
designed, static and dynamic strength requirements were mainly considered. Then, numerical
simulation and analysis of bird impact should be carried out to investigate the bird strike resistance of
the structures. During this process, material model of bird and structures should be established by
various material tests. The numerical method should be effective and credible only when it is verified
by bird strike tests. Then the simulation results could provide reasonable damage prediction and failure
mode, upon which, further structural improvements can be design. A classic flow of the process to
design a bird-strike resistant component was presented in this study as showed in Fig.1.
Fig.1. Process of design component under bird-strike Fig.2. Flow of bird material parameters optimized
The first step in anti-bird striking design is to identify the constitutive model parameters of the bird.
Bird model is very important for the numerical simulation in bird strike. The really bird body is not a
homogeneous medium. It includes bones, meat and feathers, thus, equivalent model must be
established. In addition, the bird shows different mechanical behaviors at different impact velocity. At
low speed impacting, the bird can be expressed as an elastic-plastic body while at high speed its
behavior is similar to fluid. Murnaghan Equation of State was used to simulate the bird material and the
parameters in the equation were optimized as the method showed in Fig.2.
The second step in anti-bird striking design is to identify the mechanical parameter of Aeronautical
materials at high strain rate. Bird strike usually occurs at high speed and it is indicate that the materials
experienced deformation at high strain rate. Strain rate may significantly change the material
constitutive relationship, affecting the structural response under bird strike. The constitutive model
parameters for typical aircraft structure materials were fitted in this study.
The third step is to select the reasonable numerical simulation method of bird strike process.
Currently, three simulation method-uncoupling method, coupling method and smooth particle
hydrodynamic (SPH) method are usually employed to simulate the dynamic response of the aircraft
structures under bird strike. The uncoupling method need to prior identify the bird impacting load and
then applied the load on the structure independently in numerical simulation. The coupling method
considers the simultaneous deformation for both the bird and the structure by the contact impact
coupling algorithm and fluid-structure coupling algorithm. The smooth particle hydrodynamic (SPH)
method is a meshless method capable of solving large deformation and nonlinear dynamic problems.
Technically, the SPH method is also belong to coupling method and it is used in the presented paper.
69
KEYNOTE PAPERS
According to the above method, three examples on design of aircraft structures subjected to bird
impacting load were presented in this study. For all these examples, bird strike tests were carried out on
different structures and the corresponding numerical simulation models were developed by using the
finite element software PAM-Crash. The results for tests and simulations were compared to identify
parameters of the numerical simulation models and to validate the numerical models. Finally, structural
improvements for anti-bird strike were design on these aircraft structures.
For the first example on design of anti-bird strike, a specimen component was extracted from the
cover sheet behind the aircraft radome. The design results of the component subjected to impacting load
was showed in Fig.3.
(a) Original structure (b) New designed structure

Fig.3 Deformation of the aircraft structure under bird striking for prior and post design
The second example, the performance on anti-bird strike for the vertical tail depends mainly on the
leading edge and the auxiliary beam structures. This section focuses on investigating the configuration
of the auxiliary beam on the anti-bird strike performance of the vertical tail to develop better auxiliary
beams with different configurations. The results were showed in Fig.4.
(a) Original structure (b) New designed structure-1 (c) New designed structure-2
Fig.4 Deformation of the vertical tail auxiliary beam under bird striking for prior and post design
The last example, the aircraft component tested is a representative horizontal tail structure. A new
configuration of the tail structure capable of improving the anti-bird performance was presented in the
study. Damage of the aircraft wing leading edge under bird striking for prior and post design was
showed in Fig.5.
(a) Original structure (b) New designed structure

Fig.5 Damage of the aircraft wing leading edge under bird striking for prior and post design
70

CONFERENCE ABSTRACTS
Blast, Shock and Impact Load

BLAST, SHOCK AND IMPACT LOAD ABSTRACTS

1330 - 1350 hours
Numerical and Experimental Investigation on Hail Impact on Composite

Panels
Chao Hang1a, Yang Wang2, Lei Dai1b, Zhongbin Tang1c,3 and Tao Suo1d,3
1
a
(1007476260@qq.com), b(jstxdl9@yahoo.cn), c(tangzhongbin@nwpu.edu.cn),
d
(suotao@nwpu.edu.cn)
2
Shanghai Aircraft Design and Research Institute, China (wangyang3@comac.cc)
3
Fundamental Science on Aircraft Structural Mechanics and Strength Laboratory, Northwestern
Polytechnical University, China
Hailstone impact on composite structures can result in severe damages while aircrafts are flying.
Therefore, it’s necessary to investigate the damages of high-velocity hailstone impact on composite
laminates. In this paper, three-dimensional digital image correlation (3D-DIC) method was employed to
measure the deformation field of the aluminum alloy plate under hailstone impact. By comparing the
results of experiments and simulations, the hailstone model was validated. With the validated hailstone
model, the influence of composite layup, type of reinforced fiber and matrix, impact velocity as well as
initial angle of hailstone on the damage evolution of composite laminates was numerically investigated.
Keywords: Composite laminates; Hailstone; High-velocity impact; Damage; Digital Image Correlation
74

1350 - 1410 hours
Low-velocity Impact Response of Geometrically Asymmetric Sandwich

Beams with Local Denting Effect
Jianxun Zhang, Qinghua Qina, Zhengjin Wang, Chunping Xiang and T. J. Wangb
State Key Laboratory for Strength and Vibration of Mechanical Structures,

Department of Engineering Mechanics, Xi’an Jiaotong University, China
a
(qhqin@mail.xjtu.edu.cn), b(wangtj@mail.xjtu.edu.cn)
The analytical model is developed to investigate the low-velocity impact response of geometrically
asymmetric sandwich beams. Effects of the local denting and core strength on the overall bending are
considered in the theoretical analysis. Analytical solutions are obtained for the structural response of
fully clamped geometrically asymmetric sandwich beams struck by a heavy mass with low-velocity at
midspan. Numerical results are carried out and good agreement is achieved between the analytical
predictions and numerical results. It is shown that the load-carrying and energy absorption capacities of
geometrically asymmetric sandwich structures may be overestimated if the effect of local denting is
neglected in the theoretical analysis. Also, the present quasi-static analytical solutions can reasonably
predict the low-velocity impact response of the fully clamped geometrically asymmetric sandwich
beam incorporating local denting effect.
Keywords: Geometrically asymmetric sandwich beam; Metal foam; Low-velocity impact; Local
denting; Overall bending.
75

1410 - 1430 hours
An Analytical Model for Low-Velocity Impact Response of Metal Sandwich

Beams with a Foam Core
Chunping Xianga, Qinghua Qinb, Zhengjin Wang, Mingshi Wang and T. J. Wangc
State Key Laboratory for Strength and Vibration of Mechanical Structures, Department of Engineering
Mechanics, Xi’an Jiaotong University, China
a
(xchp.09@stu.xjtu.edu.cn), b(qhqin@mail.xjtu.edu.cn), c(wangtj@mail.xjtu.edu.cn)
An analytical model for the combined local denting and overall bending deformations is developed
to predict the dynamic response of fully clamped metal sandwich beams with foam core subjected to
low-velocity impact at mid-span. Effect of the local denting on overall bending is considered in the
theoretical analysis. The membrane factor method is used to consider the effect of axial force on large
deflections of local denting and overall bending deformations. The theoretical predictions are compared
with the finite element results and good agreement is achieved. It is shown that the sandwich beam is
strengthened with the deflection increasing.
Keywords: Metal sandwich beam; Low-velocity impact; Local denting; Overall bending; Large
deflection; Analytical model.
76

1430 - 1450 hours
Development of a New Class of Penetration Analysis Methods

Youcai Wua, John E. Crawfordb, Hyung-Jin Choi, Shengrui Lan and Tim Brewer
Karagozian & Case, Inc., United States

a
(wu@kcse.com), b(crawford@kcse.com)
A new class of penetration analysis methods has been developed by Karagozian & Case, Inc.
(K&C). The method developed represents a combination of RKPM (reproducing kernel particle
method) and FEM (finite element method) formulations, where the coupling is evolutionary in nature.
Results for small and large projectiles (i.e., 1 to 1,000 kg) are shown for a range of impact velocities.
Results from LS-DYNA are also shown illustrating the major discrepancies exhibited by conventional
coupled methods in performing penetration analysis.
Keywords: Impact; Penetration; Analysis; Numerical simulations
77

1450 - 1510 hours
Environmentally Low Impact Tunnel Blasting in Urban Residential Area

Keita Iwano1a, Takaaki Inuzuka1b, Kinya Kuriki2, Takayuki Ochikawa3,
Katsunori Fukui3 and Hiroshi Andoh4
1
Rock Mechanics and Hydrogeology Group, Kajima technical Research Institute, Japan
a
(iwanokeita@kajima.com), b(inuzukat@kajima.com
2
River division, Construction Department, Japan (kasen@city.toyota.aichi.jp)
3
Department of Systems Innovation, University of Tokyo, Japan (fukui@sys.t.u-tokyo.ac.jp)
4
Orica Japan Co. Ltd., Japan (hiroshi.andoh@orica.com)
Reducing the environmental impact caused by tunnel blasting is quite important. Aneigawa tunnel,
which runs under the residential area with its geological condition of quite hard granite, was required to
make a low vibration to the residential area. New electronic detonator; eDevII, resolved the problems
with highly accurate delay time control. Authors made a series of blasting test with this detonator, such
as properties of blasting vibration waveforms, frequency characteristics of vibration, relation between
distance and peak vibration, and we obtained a lot of data and know-how to control blasting vibration at
the highly residential area. And we also introduced new vibration prediction method (MSW: Multi Seed
Waveform) to this tunnel and tried to compared real waveforms with predictions.
Keywords: Tunnel blasting; Electronic detonator; Delay time; Vibration prediction
78

1540 - 1600 hours
Local Damage of Fiber Reinforced Cementitious Composite Plates

Subjected to High Velocity Impact
Masuhiro Beppu1a, Hironori Ueno1b and Atsuhisa Ogawa2
1
Department of Civil and Environmental Engineering, National Defense Academy, Japan
a
(beppu@.nda.ac.jp), b(em52054@.nda.ac.jp)
2
Fibers and Industrial Material Division, Kuraray Co. Ltd., Japan (atsuhisa_ogawa@kraray.co.jp)
In recent years, explosive incidents due to terrorist’s bombings have increased in the world. Local
damage of concrete structures occurs frequently due to secondary fragment impacts by the explosive
loads. Hence, it is necessary to establish a reinforcement method that is able to take account of the high
velocity impact effects for vital structures such as a nuclear power plant facility. This study examined
the local damage resistant performance of fiber reinforced cementitious composite plates with
polypropylene or polyvinyl alcohol fibers subjected to high-velocity impact by a rigid steel projectile.
In a series of tests, a steel rigid projectile with a mass of 46g was used for a projectile and three types of
fiber reinforced cementitous composite plates were investigated. To examine failure mode of the
specimen, impact velocity varied between 200m/s and 400m/s and the impact tests were conducted until
penetration occurred. The penetration and scabbing limit thicknesses were estimated by multiplying a
reduction coefficient by the modify NDRC formula. In addition, high velocity impact tests of the PVA
fiber reinforced cementitious composite plates with fiber mix ratios of 1.0vol% and 2.0vol% were
conducted to examine restraining effects by the fiber mix ratio.
Keywords: Fiber Reinforced Cementitious Composite; Local Damage; High Velocity Impact
79

1600 - 1620 hours
Impact Resistance of Sand against Projectile Penetration

Siau Chen Chian1, Beng Chye, Vincent Tan2 and Sarma Anand3
1
Department of Civil and Environmental Engineering, National University of Singapore, Singapore
(sc.chian@nus.edu.sg),
2
Department of Mechanical Engineering, National University of Singapore, Singapore
(mpetanbc@nus.edu.sg),
3
Defence Science & Technology Agency, Singapore (sanand@dsta.gov.sg)
The use of sandbags as a ballistic impact resistance barrier is conventionally used for military field
fortifications. Compacted granular material, like sand, has a tendency to dilate under shear loading. This
occurs when there are low void spaces between the sand grains and the interlocking grains do not have
the freedom to rearrange themselves to a denser state. Such tendency of dilation is high if the relative
compaction of the sand sample is high. Furthermore, the larger the strain rate or loading rate, the higher
is the peak strength of the sand sample. A suite of impact tests were conducted with the smooth bore gas
gun at the National University of Singapore. Different shapes and mass of projectiles were fired into
sand specimens with varying relative densities. High speed cameras were used to determine the impact
and exit velocities of the projectile during the testing. This paper discusses the ballistic performance of
sand over a wide range of variables such as relative compaction of sand, impact velocity, mass and
shape of projectile.
Keywords: Ballistic impact; Sand; Projectile; Dilation
80

1620 - 1640 hours
Single Impact and Residual Strength Tests on Two-way Concrete Panels:

Effects of Adding Steel Fiber and Changing Rebar Detail
Young-Soo Yoon1a, Ki-Yeon Kwon2, Jin-Young Lee1b and Hyun-Hoon Choi3
1
Environmental and Architectural Engineering, Korea University, Korea
a
(ysyoon@korea.ac.kr), b(jinyounglee@korea.ac.kr)
2
Structural System & Site Evaluation Department, Korea Institute of Nuclear Safety, Korea
(kiyeon83@gmail.com)
3
Building R&D, Samsung C&T, Korea (hh2009.choi@samsung.com)
The impact resistances of two-way concrete panels were studied by conducting the single impact
test and the residual strength test. In the single impact test, the impact load was applied to each
specimen only one time by dropping weight on the specimen from a certain height. The applied impact
load level was adjusted by changing the mass of the weights from 300kg to 400kg. The drop height was
fixed as 2,000mm. The performances of the specimens under the single impact load were evaluated by
comparing their reaction force, deflection, and surface damage. In the residual strength test, the load
was increased statically maintaining the same boundary condition with the single impact test. The
residual strength test was stopped when the measured applied load started to decrease. The
performances of the specimens under the static load were evaluated by comparing their strength, and
toughness values. Based on the single impact and the residual strength test, it was observed that the
resistance against impact load can be improved by applying dense reinforcement layout and fabricating
using fiber reinforced concrete. Moreover, these changes (dense reinforcement layout and fiber
reinforced concrete) can also lead that less reduction in strength and toughness of the specimen after the
specimen was subjected to the single impact load.
Keywords: Single impact; Residual strength; Dense reinforcement layout; Fiber reinforced concrete;
Two-way concrete panel
81

1640 - 1700 hours
Energy Absorption Performance of Ni-Based Superalloy Honeycomb

Structure under Different Ball Velocities
Kang-Bo Yuana, Wei-Guo Guob and Xue-Ming Tanc

a
(kangbo0528@163.com), b(weiguo@nwpu.edu.cn), c(124841817@qq.com)
In this study, to understand energy-absorbing performance of one kind of honeycomb structure of

Ni-based superalloy, uniaxial compression tests, ballistic impact tests, and numerical analysis are
performed. The compression experiments are conducted over a strain rate range of 0.001/s-1500/s and a
temperature range of 20º -700º using Split Hopkinson Pressure Bar and quasi-static testing machine.
Gas gun is used for the impact tests under different temperatures (20º and 600º), impact speeds (50m/s,
150m/s, 275m/s) and impact angles(0º, 45º and 75º) of the projectiles. The results show that the
compression property of the honeycomb structure is more sensitive to temperature at high strain rates
than at high strain rates. The crater depth increases linearly with the increasing impact speed, and the
increasing rate decreases with the increasing impact angle. Through metallographic analysis on the
microstructure of the honeycomb structure, a real structural finite element model of the honeycomb
panel is built for the numerical finite element impact tests. The numerical simulation results are good
agreement with experimental results.
Keywords: Metal honeycomb; Energy absorption; Ni-based superalloy; SHPB, Gas gun
82

1700 - 1720 hours
Optimization of the Material Systems with Magnetorheological Fluids

Adam Wisniewski1*, Dawid Pacek1, Joanna Kozłowska2, Paweł Zochowski1, Marcin
Leonowicz2 and Lukasz Wierzbicki2
1
Military Institute of Armament Technology, Poland
*
(adam.wisniewski@witu.mil.pl)
2
Faculty of Materials Science and Engineering, Warsaw University of Technology, Poland
(jkozlowska@inmat.pw.edu.pl)
The aim of present study was to develop the ballistic inserts for the new generation of modular
lightweight bullet-proof vests of high ballistic resistance and large surface of the protected human body,
which could provide greater mobility in comparison to currently used vests.
The scope of research covered ballistics tests of 100x100 mm samples containing
magnetorheological fluid (MRF), fired with the 9 mm Parabellum FMJ bullet. Behaviour of samples in
and out of a magnetic field was investigated and optimization of composite structures was done.
Various magnetorheological fluids were used for the tests. All fluids were composed of synthetic
oil, magnetic particles in a form of carbonyl iron, thixotropic agent and stabilizers.
Different kinds of bags were filled with the MRFs: bags made of rubberized aramid or polyamide
layers, bags made of non-woven UHMWPE Dyneema ® SB71, polyethylene zip-bags, and silicone
bags.
Beside the bags filled with MRF also MRE elastomers and MRE elastomers filled with MRF were
used. All the samples contained aramid and / or polyethylene (UHMWPE) textile layers.
The paper describes what effect the textile, type and volume of MRF, variant of the MRF use and
the bag location have on protective properties of samples, expressed by backface signature (BFS) – the
greatest extent of indentation in the backing material caused by a non-perforating impact on the armour.
It was found that addition of MRF and reduction of number of textile layers in the composite
material system increases its flexibility without decreasing protective properties. In case of rubberized
bags, 35% reduction of textile layers number in comparison to the “neat” layers was obtained.
Keywords: Armour; Bullet-proof vest; Magnetorheological fluid; Magnetorheological elastomer;

Textile
83
Wednesday, 20th May 2015

1050 - 1110 hours
Numerical Predictions of the Negative Phase

Piotr Sielicki1 and Samuel Rigby2
1
Poznan University of Technology, Poland (piotr.sielicki@put.poznan.pl)
2
University of Sheffield, United Kingdom
The field of blast protective design emerged in the late 1940s and focussed mainly on large scale
(nuclear) explosive loading massive structures. In these situations, positive phase effects were seen to
dominate and the negative phase could effectively be ignored. Recently, however, the threat has moved
to smaller scale explosives and increasingly lightweight structures. Here, the negative phase becomes
important, however despite this the negative phase is often overlooked.
This research presents a numerical investigation on the negative phase, with a primary focus on an
accurate numerical scheme for modelling the negative phase blast pressure. Numerical tests are
performed on deformable targets to determine fully reflected blast parameters, with associated
numerical modelling conducted using Abaqus/Explicit. Moreover, the failure modes are obtained for
light-weight panel employing the Perzyna model for metallic materials.
The computational methods are adapted for better representation of the negative phase, including
mesh refinement strategies, modelling of the explosive event and accurate description of the air
behaviour. The results herein can be used to inform blast resistant designers on how to accurately model
negative phase effects.
Keywords: Blast loading; Negative phase; EOS modeling
84

1110 - 1130 hours
Blast Load Assessment for Safety in Urban and Close-in Environments

David Stevensa, Matt Barsottib and Carrie Davisc
Protection Engineering Consultants, United States

a
(dstevens@protection-consultants.com), b(mbarsotti@protection-consultants.com),
c
(cdavis@protection-consultants.com)
The pressure histories generated by explosive detonations within urban and close-in environments
pose a significant hazard to personnel, including law enforcement, first responders, bystanders and
hostages. Two recent projects performed by the U.S. Department of Defense (DoD) and the Department
of Homeland Security (DHS) have addressed the prediction of blast pressure histories in these
environments. In the first project, numerical simulations were performed to develop a simple safe
standoff tool and training animations for law enforcement personnel who may be subjected to explosive
overpressure during render-safe operations of improvised explosive devices (IEDs) and explosive
breaching of structural components (doors, walls, windows, etc.). These animations and tools are
implemented within new training courses that will improve the ability of first responders and law
enforcement personnel to identify safe areas during explosive operations. In the second project,
propagation of shock in an urban environment was studied experimentally and numerically. Critical
data on the reflection and diffraction of blasts at large scale were created; this data was used to assess
the accuracy of state-of-the-art numerical codes for predicting blast propagation. The effects of
accuracy in load predictions on the type and cost of building hardening and on public safety were also
examined.
Keywords: Blast; Pressure histories; Explosive breaching; Bomb detonation; Urban environment; First
responders.
85

1130 - 1150 hours
Blast Effects inside a Partially-Enclosed Car Park

Choon Keat Anga, Wei Ping Chengb and Jing Yan Kongc
Prostruct Consulting Pte Ltd, Singapore
a
(choonkeat.ang@prostruct.com.sg), b(wp.cheng@prostruct.com.sg), c(jy.kong@prostruct.com.sg)
Modern buildings are commonly designed with densely populated offices or residential occupants
above several storeys of car parks below. In case of an unforeseen explosive threat that occurred at the
car parks, such buildings should protect the occupants from a catastrophic progressive collapse.
Modelling and prediction of blast effects inside a partially-enclosed space due to such threats are
therefore of interest to building designs. In this paper, the blast effect inside a partially-enclosed car
park subject to a vehicle-borne improvised explosive device is modelled using a commercially available
Computation Fluid Dynamics (CFD) code. The corresponding responses of the primary structural
elements inside this enclosure due to the blast loads are analyzed to confirm that the car park is able to
withstand this blast load and protect the occupants above from such a threat.
Keywords: Partially-enclosed; Car park; Blast overpressure; Numerical simulations
86
Thursday, 21st May 2015

1150 - 1210 hours
Coupled CFD/CSD Simulations of Dust Production by Fragmenting

Charges
Joseph D. Baum1a, Orlando A. Soto1, Fumiya Togashi1, Rainald Löhner2,
Robert A. Frank3, Ali Amini1 and Jaqueline Bell1b
1
Applied Simulations, Inc. USA
a
(joseph.d.baum@appliedsimulations.com) ,b(Jackeline.Bell@dtra.mil)
2
George Mason University, USA (rlohner@gmu.edu)
3
Applied Research Associates, Inc, USA (rfrank@ara.com)
We present the results of coupled Computational Fluid Dynamics (CFD)/Coupled Structural Dynamics
(CSD) simulations of internal detonations of cased munitions against reinforced concrete walls. These
simulations are part of a test, analysis, and modeling effort studying air blast propagation through breached
walls. The coupled CFD/CSD simulations are providing additional insight and details not measured in the
tests, as well as developing a synthetic database to supplement the test matrix.
Initial simulations are being performed to calibrate the CFD/CSD model and to determine the physic
impacting the internal environments, wall breach, and blast propagation through the breach. The initial
simulation modeled the response of two reinforced concrete walls to loads from a cased charge, placed in
close proximity to the center of wall 1.
In the test, the detonation room (composed of two culverts) incurred a large amount of damage in the
test due to the fragments and blast load, and both culverts failed. Test wall 1 initially breached over the
middle third, with the wing walls removed by the later time blast loads. Debris from test wall 1 impacted test
wall 2, which failed under the combined blast and debris.
Initial coupled CFD/CSD simulations modeled the culverts as rigid, non-responding surfaces. These
simulations reproduced the damage to the test walls, but the pressure histories matched the experimental data
only out to 10 ms. Subsequent airblast reflections were significantly reduced, as if a large amount of energy
has been evacuated from the facility. Post-test damage analysis showed significant fragment damage to the
culverts, with the concrete stripped to the first layer of rebars. We estimate that the fragment impacts
produced several hundred kilograms of dust that was ejected into the room. Repeat simulations, where the
culvert response was modeled and the dust was allowed to absorb both kinetic and thermal energy, matched
the experimental data significantly better.
Additional simulations are being performed to further validate our modeling approach and to provide a
synthetic database for fast running model development (FRM). FRMs being developed or updated include
opening creation (size and timing), secondary debris, and blast propagation through the breach opening.
Keywords: ICFD; CSD; Airblast; Breach; Blast propagation; Dust; Cased munitions impact; Impulsive;
Shock; Protective design; Numerical simulations
87

1210 - 1230 hours
Approach to Combined Blast and Fragmentation Modelling

Ke Wei, Joel Ong*, Hui Wei Yeo and Oi Yin, Karen Chong
Defence Science & Technology Agency, Singapore

*
(okewei@dsta.gov.sg)
A modelling approach to combined blast and fragmentation analysis is outlined. Validation against
past momentum plate tests has been performed. The findings indicate that the model is able to predict
the blast and fragment loading on a non-responding surface with good agreement with experimental
results. Accurate modelling of the casing behaviour during the initial expansion phase was found to
have significant impact on the subsequent expansion of detonation products and air blast propagation. It
was found that it is also important to account for the smaller mass bins in Mott’s distribution when
initializing the fragment mass distribution. Simulations were performed to derive trends on varying
casing thickness, standoff distance and time of arrival.
Keywords: Numerical simulation; Blast load; Fragment impact
88
Impact and Explosion Tests

IMPACT AND EXPLOSION TESTS ABSTRACTS

1330 - 1350 hours
Fibre Reinforced Composite Piping Systems against Impact

D. X. Xionga, T. Y. Wangb and J. Y. Richard Liewc

a
(ceexd@nus.edu.sg), b(ceewty@nus.edu.sg), c(ceeljy@nus.edu.sg)
As shallower oil wells become depleted, exploration and production will move increasingly
towards offshore and deeper water, where conventional fix platforms and jack-up rigs which are made
of steel are no longer feasible and the request for deep water structures that are made of lighter and
stronger material is growing. Fibre reinforced plastic (FRP), which has demonstrated its advantages in
terms of lightweight, high strength, and high corrosion resistance in automobile and aeronautic industry,
will be an alternative solution to deep-water riser and pipeline systems which are vulnerable to
accidental impact loads. This paper presents an investigation on the behaviour of hollow fibre
reinforced plastic (FRP) pipes subjected to local transverse impact. Both of glass FRP (GFRP) and
carbon FRP (CFRP) pipes were tested. Drop weight impact tests were adopted and the impact projectile
was freely released at different drop heights. Different failure modes were observed and the critical
factors affecting damage were discussed. The performance of GFRP and CFRP pipes under lateral
impact was assessed for potential application in deep-water piping systems.
Keywords: Impact; Composite materials; Composite riser; GFRP; CFRP
92

1600 - 1620 hours
Small-Scale Vapour Cloud Explosion Tests

Yadong Lina and Chor Boon Ngb

a
(lyadong@dsta.gov.sg), b(nchorboo@dsta.gov.sg)
Flammable, explosive gas when mixed in air forms a vapour cloud mixture that can ignite and
explode to release a large amount of energy when certain conditions are met. The characteristics of
vapour cloud explosives (VCE) are different from conventional explosives in that VCEs are capable of
generating blast waves with longer duration and higher temperatures. The prediction of blast effects
generated by VCEs can be grouped into three categories: TNT equivalence method, numerical
simulations and blast curve methods. A number of small scale VCE tests had been conducted using
Propylene Oxide (PO) as the fuel. The tests were conducted near the opening of a shock tube to
understand the effects of VCEs in open environment and semi-confined environment. The paper
describes the considerations for conducting a VCE test and will share how the test results compare with
the predictions from TNT equivalence method and blast curves.
Keywords: VCE; Blast effects; TNT equivalence; Blast curve methods; Baker-Strehlow-Tang (BST)
method; TNO multi-energy method
93

1410 - 1430 hours
Explosive Trial to Investigate the Hazard Mitigation Effects of Earth Cover

Kang Kok Wei1a, Knut Gaarder Rakvåg2 and Chua Hiong Suan1b
1
Defence Science and Technology Agency, Singapore
a
(kkokwei@dsta.gov.sg), b(chiongsu@dsta.gov.sg)
2
Norwegian Defence Estate Agency, Norway (Knut.Gaarder.Rakvag@forsvarsbygg.no)
An earlier series of tests on small, 1/5th scale earth covered magazines (ECM),showed that the earth
cover can reduce debris generated after explosives stored in reinforced concrete (RC) structures
detonate. The explosives are part of the ammunition stored in the RC buildings. Their detonation arises
from accidents. In August 2014, another two RC box structures with explosives within were tested.
However this time they were built to a larger 2/5th scale. Each box had internal dimensions of 2x2x1.6m.
Each box stored a net explosive quantity (NEQ) of 128kg. However, the two structures had different
thicknesses of earth cover; 0.48m and 0.96m thick covers over each box’s RC roof. The detonation
events were monitored using pressure gauges and high speed cameras. After each detonation, concrete
and steel rebar debris were collected along principal axes to assess the hazards emanating from the
explosions within the boxes. Onsite observations after the detonations showed that additional earth
cover reduced the damage to the structure which had housed the detonating explosives. High speed
video footage of the thicker earth covered structure also showed that other than the front sector which
included the door and which had no earth cover, most of the debris from the box breakup was launched
almost vertically upwards. This resulted in a significant reduction of debris at the side and rear sectors.
This observation was quantified by observations and measurements of the distributions of both debris
numbers and debris density for both detonation events. The pieces of debris which landed furthest in
both detonation events were the doors. These were located more than 800m from the structure. Other
significant pieces of debris were also was found from the sectors in front of the exploding test
structures. These front sectors had not had earth covers, and were where the doors were. These other
pieces of debris were from portions of wall above the doors of the structure. In both detonation events,
this portion of the front wall landed first in one piece at around 540m from the structure, and then broke
up upon impact with the ground, generating more pieces of debris. The effect of earth cover to mitigate
blast pressure levels and distribution was less pronounced than earth cover’s effect on debris.
Preliminary comparisons with the smaller scaled ECM are also discussed in this paper.
Keywords: Earth Covered Magazine; Blast; Debris; Hazards; Explosive Storage
94

1430 - 1450 hours
Observations from Blast-Induced Liquefaction Tests in Singapore

Anand Sarma*, Hiong Suan Chua and Eu Chang, Leonard Heng

*
(sanand@dsta.gov.sg)
The susceptibility of saturated sandfill to blast-induced liquefaction was investigated by

undertaking a field explosion trial, complemented by laboratory tests and numerical modelling. While
many researchers have studied liquefaction mechanisms and factors influencing liquefaction from
earthquake studies, there is a dearth of publications on blast-induced liquefaction. Interest in
blast-induced liquefaction was initiated mainly in the 1980s to understand liquefaction potential due to
concentrated loads. Recent trials using distributed arrays in New Zealand have demonstrated that the
use of buried explosive charges is an effective method of modelling earthquake events, including
ground response and coupling effects with buildings. This paper addresses three areas related to
blast-induced liquefaction. First, the conduct of a field explosion trial incorporating subsurface
detonations and free-field instrumentation measurements is described in detail. The design of a
placement system for sub-surface data acquisition is also described. Second, preliminary analyses of
field records from the resulting residual pore pressure ratio is discussed and used to estimate the zone of
influence of liquefaction due to the detonations. Thirdly, the zone of influence is compared with
quasi-empirical predictions suggested by other researchers and conclusions drawn.
Keywords: Blast-induced liquefaction; Field trial; Pore pressure ratio; Zone of influence
95

1450 - 1510 hours
Investigation of Dynamic Deformation Behaviors in Structural Steels at

Intermediate Strain Rates using Newly Constructed Impact Tensile Test
Apparatus
Hyung-Seop Shin1*, Kyung-Oh Bae1, Lee-Ju Park2 and Hyung-Won Kim2
1
Department of Mechanical Design Engineering, Andong National University, South Korea
*
(hsshin@anu.ac.kr)
2
Agency for Defence Development, South Korea (leejupark@yahoo.co.kr)
Some steel structures for the purpose of protection and safety are usually subjected to a dynamic
load corresponding to the intermediate strain rate of ~500 s-1. Therefore, it is necessary to evaluate the
load bearing/prevention capability in those cases. However, the intermediate strain rate region is located
between quasi-static and high strain rate region utilizing SHPB testing apparatus. Therefore, it is not
easy to obtain the test condition using any conventional apparatus. In order to solve the difficulty, we
have constructed a dynamic tensile testing device working at intermediate strain-rate region which
utilizes a long rod and cylinder as an output bar and drop-bar weight, respectively. Using the newly
constructed impact tester, the dynamic deformation behaviors of various structural steels in the
intermediate strain rate region were investigated and compared with the quasi-static characteristics.
Keywords: Impact tensile test; Intermediate strain rate; Long output bar; Drop-bar; Structural steel
96

1540 - 1600 hours
Experimental Study of Non-Composite SCS Sandwich Panels under

Impulsive Loading
Yonghui Wang*, J. Y. Richard Liew and Siew Chin Lee

*
(wangyonghui@nus.edu.sg)
The Steel-Concrete-Steel (SCS) sandwich panel consists of a concrete core connected to two steel
faceplates using mechanical shear connectors and it was proven with high blast resistance capacity and
prior to the traditional concrete and steel panels in terms of high ductility, spalling protection and
buckling resistance. Hence, several researches were conducted to study the static, impact and blast
performances of SCS sandwich panels with different types of shear connectors. However, little research
has been done on the SCS sandwich panels without shear connectors. Therefore, the aim of this research
is to study the non-composite SCS sandwich panels under impulsive loading. Two SCS sandwich
panels with different core depths were tested and the impulsive loading was achieved in the laboratory
by utilizing an inflated high pressure airbag to transfer the applied load from dropped projectile onto the
specimens. The deformation modes, applied pressures, displacement and strain responses were
extracted from the tests to study the behaviors of non-composite SCS sandwich panels under impulsive
loading. The Finite Element (FE) method was also adopted to simulate the tests and reasonable
agreement between the two was observed.
Keywords: Steel-Concrete-Steel; Impulsive loading; Failure mode; Numerical study
97

1600 - 1620 hours
Blast Tests and Finite Element Simulations on Sandwich Beams

with Tubular Cores
X.M. Xiang1a, G. Lu2, Xiangyu Li3 and D.W. Shu1b
1
School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore
a
(xian0005@e.ntu.edu.sg), b(MDSHU@ntu.edu.sg)
2*
Faculty of Science, Engineering and Technology, Swinburne University of Technology, Australia
(glu@swin.edu.au)
3
College of Science, National University of Defense Technology, China (xiangyu_lee@163.com)
This paper presents experimental investigations and finite element simulations of sandwich beams
with tubular cores subjected to transverse blast loads. The beam cores consist of closely arranged
circular or rectangular tubes. Different beams core configurations, consisting of varying tube types,
varying tube sizes, and varying tube spacing, were studied. The cores were subjected to different masses
of TNT at various standoff distances from the beams. The blast loads were sufficient to cause plastic
local deformation of the thin-walled tube cores and global plastic flexural deformation of the whole
beams. LS-DYNA was used for FEA simulations and the results were compared to the experimental
profiles of global and local deformations. In the FEA simulations, the full lengths of the beams were
modeled using an elastic-plasticity material model with strain hardening.
Keywords: Sandwich beams; Sandwich beams; Blast tests; Simulation
98

1620 - 1640 hours
Experimental Investigation of Response of Steel Plates due to Close-in Blast

Loading from Spherical Liquid Explosive
Alex Remennikov1a, Tuan Ngo2a, Damith Mohotti2b, Brian Uy3 and Michael Netherton1b
1
Centre for Infrastructure Protection and Mining Safety, University of Wollongong, Australia
a
(alexrem@uow.edu.au), b(michael.netherton@newcastle.edu.au)
2
Department of Infrastructure Engineering, The University of Melbourne, Australia
a
(dtngo@unimelb.edu.au), b(pmdamith2012@gmail.com)
3
Centre for Infrastructure Engineering, School of Civil and Environmental Engineering,
The University of New South Wales, Australia (b.uy@unsw.edu.au)
Detonations of nitro methane spherical charges have been carried out to study near-field blast
loading of steel plates and the effectiveness of several protective solutions. Three types of bare steel
plates, namely mild steel, high-strength steel, and stainless steel were subjected to explosive blast
loading. Steel plates of the same type with polyurea coating and composite covers were also subjected
to localised blast loading. During an explosive field trial, the blast pressures and displacements of steel
plates were measured. Additionally, loading of steel plates by the impinging detonation products was
captured by high-speed video recordings. This experimental program has produced results which can be
used to calibrate numerical models and to refine the simplified models for predicting blast loads and
response of structural elements due to close-in detonations. The effectiveness of polyurea coatings and
other solutions for enhancing blast protection of vehicle and ship structures is discussed.
Keywords: Impulsive loading; Near-field blast; Protective design; Liquid explosive
99

1640 - 1700 hours
Lesson from Blast Test of Steel Plate in Elastic Domain

Jung Hun Kee1a, Young Hak Lee2a, Eunsun Jo2b, Seung Jae Lee3 and Jong Yil Park1b
1
Department of Safety Engineering, Seoul National University of Science and Technology, South Korea
a
(junghunkee@gmail.com), b(jongyilpark@gmail.com)
2
Department of Architectural Engineering, Kyung Hee University, South Korea
a
(leeyh@khu.ac.kr), b(joeunsun@khu.ac.kr)
3
Agency for Defense Development, The 8th R&D Institute, South Korea (banana76@add.re.kr)
To validate and improve Single-Degree-of-Freedom (SDOF) method, steel plates were tested
under explosion. Though AUTODYN provided good estimation comparing test results, SDOF results
had significant difference. Tension membrane was not considered in SDOF, since plates were design to
behave in elastic domain. AUTODYN model excluding tension membrane effect shows good match
with SDOF results. Development of SDOF with tension membrane in elastic domain is recommended.
Keywords: Test; Steel Plate; AUTODYN; Explosion; SDOF
100

1700 - 1720 hours
Photonic Doppler Velocimetry Instrumentation in Real Scale Concrete Slab

Blast Loading Experiments with an Attempt to Measure the Influence of the
Type of Concrete on the Debris Velocity
Martin Künzel1, Marek Foglar1 and Jiri Pachman2
1
Department of Concrete and Masonry, Czech Technical University in Prague,
Czech Republic (marek.foglar@fsv.cvut.cz)
2
Institute of Energetic Materials, Faculty of Chemical Technology, University of Pardubice, Czech
Republic (jiri@pachman.eu)
The author team has performed a set of experiments in the past years focused on blast performance
of concrete slabs subjected to adjacent blast loading. The 2014 tests focused on blast performance of
slabs with 40 or 80 kg.m-3 of steel fibers; the fibers were high strength, fy = 2000 MPa. The slabs made
of UHPFRC were tested, one of the specimens with a glass-fiber matrix inserted in the concrete cover.
The velocity of the debris at the soffit of the slabs under the blast loading was recorded using
photonic Doppler velocimetry. The paper presents the experimental data and discusses them.
Keywords: Blast loading; Concrete; Photonic Doppler velocimetry
101
Material Behaviour under Blast

MATERIAL BEHAVIOUR UNDER BLAST ABSTRACTS

1050 - 1110 hours
A Micromechanical Method to Predict Macroscopic Behavior of Progressive

Failure in Brittle RockM
Xiaozhao Li and Zhushan Shao*
School of Civil Engineering, Xi’an University of Architecture & Technology, China

*
(shaozhushan@xauat.edu.cn)
Microcracks growth has great influence on the progressive failure of brittle rock. A major
challenge in this area is to link microcrack growth with macroscopic mechanical behavior. Based on the
macroscopic and micromechanical definition of damage in rock, a new model is suggested to establish
the linkage between microcrack length and macroscopic strain. Using this model, the progressive
failure behaviors are studied by employing the Ashby and Sammis’ microcrack model in brittle rock.
Reasonability of the suggested model is verified by experimental results.
Keywords: Brittle rock; Micromechanics; Crack growth; Progressive failure; Damage
104

1110 - 1130 hours
Reaction Force and Delamination Properties of Laminated Glass Windows

Subject to Blast Loads
Paolo Del Linz1*, Yi Wang2, Paul A. Hooper2, David Smith3, David Cormie1,3, Luke
Pascoe3, Bamber R. K. Blackman2 and John P. Dear2
1
2
Imperial College London, Department of Mechanical Engineering, United Kingdom
3
Arup Resilience Security and Risk, United Kingdom
*
(paolo@ntu.edu.sg)
In this research performed at Imperial College London, blast tests have been carried out to measure
the deformation of realistic glass panes under full scale blast conditions. Data was collected using a
combination of strain gauges and Digital Image Correlation techniques (DIC). This information,
complemented with finite element analysis (FEA), has been used to firstly calculate the reactions
produced in a real blast both before and after the glass layers failure. Laboratory tests performed on the
Polyvinyl Butyral (PVB) material and on pre-cracked laminated glass specimens were then employed
to calibrate FEA models with the aim of estimating the delamination energy of this material.
Keywords: Blast; Laminated glass; Delamination; Viscoelasticity; Numerical simulations
105

1130 - 1150 hours
Damping Characteristics of Sand-Rubber Chip Mixtures as a Seismic

Barrier
Zhuoyuan Chenga and Eng Choon Leongb

a
(zcheng003@e.ntu.edu.sg), b(cecleong@ntu.edu.sg)
Rubber tyres reached the end of their life regularly and enter the waste cycle. The large volume of
disposed rubber tyres can create a negative impact on the environment if they are not disposed properly.
As such, various initiatives have been proposed to reuse the rubber tyres so as to minimise on the waste
generated. One such innovative uses of waste rubber types is to take advantage of their high damping
capacity and mix them with soils to shield structures from seismic or impact loadings. In order for this
to be achieved, dynamic properties (small strain stiffness and damping ratio) of the mixtures have to be
ascertained so that they can be incorporated into the design process. In this paper, determination of these
properties, especially damping ratio, were performed in a triaxial setup using bender-extender elements.
The P and S wave velocities were determined using the travel-time method. Their corresponding
damping ratios were determined by the Hilbert transform method. Sand-rubber chips mixtures of
various proportions were tested at different effective confining pressures. The experiments highlighted
the capabilities of using signals from bender-extender elements to determine the damping ratio of a
soil-rubber chip mixture. Damping ratios of the sand-rubber chip mixtures increase with increasing
rubber chip content and decrease as the effective confining pressure increases. The damping properties
of the sand-rubber chip mixtures compared favourably with existing materials used for similar
purposes.
Keywords: Soil-rubber mixture; Damping ratio; P and S wave velocity; Hilbert transform
106

1150 - 1210 hours
Strain Rate Effects of Compressive Properties of Hollow Glass

Microspheres Filled Polymers
Xin Zhang1, Pengfei Wang2a, En-Hua Yang2b and Jinglei Yang2c
1
Interdisciplinary Graduate School, Nanyang Technological University, Singapore
(XZhang024@e.ntu.edu.sg)
2*
a
(PFWang@ntu.edu.sg), b(EHYang@ntu.edu.sg), c(MJLYang@ntu.edu.sg)
In this study, quasi-static and dynamic compressive tests were carried out to investigate the
mechanical properties of the Hollow Glass Microspheres (HGM) filled polymer with different volume
fractions of glass microspheres from 0% to 20% by volume. The strain rates applied in compressive
tests range from 0.0005s-1 to 2,750 s-1. It has been found that the HGM filled polymer may absorb more
energy at low volume fraction under low strain rate compression. The compressive modulus increases
with the increase of strain rate when the specimens were tested under low strain rates. Surface treatment
of glass microsphere provides marginal enhancement in compressive properties. Most importantly, the
HGM filled polymer exhibits strong strain rate sensitivity which decreases with the increase of volume
fraction of glass microspheres. An empirical equation was established to predict the compressive
strength of the HGM filled polymer as a function of strain rate and volume fractions of glass
microspheres.
Keywords: HGM filled polymer; Strain rate sensitivity; Empirical equation
107

1210 - 1230 hours
Design and Analysis of Protective Structures: A Fresh Look at Concrete

Peter W Weber
ceEntek Pte Ltd, Singapore

(pww@ceentek.com)
Most of the focus for Design and Analysis of Protective Structures has been on the architectural
and construction side. The base materials, concrete and steel have not changed much and are taken as a
given. New materials like ultrahigh performance concrete (UHPC) have not managed to break into the
market based on cost, performance, complexity of system, handling and special curing requirements. In
addition the overall use of steel has further increased as these materials typically use steel fibers to reach
the required performance characteristics.
The use of carbon nanofibers in UHPC allows a fresh look at UHPC systems making them much
simpler by reducing the number of ingredients required and eliminating the need for steam curing.
While not all issues have been solved at this point in time a stable, easy to prepare and use UHPC
system has been developed meeting the most of the requirements for Protective Structures. In addition,
a foamed version of UHPC can be used to optimize energy absorption leading to new composite
construction elements.
Keywords: Concrete; UHPC; Foam UHPC; Carbon nanotubes; UHPC based composites
108
Numerical Modelling and Simulation

NUMERICAL MODELLING AND SIMULATION ABSTRACTS

1050 - 1110 hours
Numerical Study on Ballistic Impact of Perforated Armour Plates

Qian Shi, Qinghua Qina, Weilong Ai and T. J. Wangb
a
(qhqin@mail.xjtu.edu.cn), b(wangtj@mail.xjtu.edu.cn)
In this paper, 3-D numerical models are developed to investigate the projectile impacts against the
perforated armor plates. Effects of impact position, obliquity, strain hardening and strain-rate are taken
into account in finite element analysis. It was found that the high ballistic resistance is achieved by
arresting the propagation of cracks in high hardness perforated plates, controlling the deflection of
projectiles and generating the high stress in the projectile. The ballistic resistance of the perforated
add-on armor is sensitive to the impact position. The oblique impact diminishes the penetration depth
especially at a large oblique angle, while it is not obvious for the angles less than 45º.
Keywords: Ballistic impact; Perforated plate; Geometry property; Obliquity and position; Finite
element method
112

1110 - 1130 hours
Design and Optimization of Laminated Transparent Armor by Finite

Element Analysis
Yukolthorn Sriloy1, Ryan McCuiston2 and Rojana Pornprasertsuk1,3
1
Research Unit of Advanced Ceramics, Department of Materials Science, Faculty of Science,
Chulalongkorn University, Thailand (yukol.sriloy@gmail.com)
2
Department of Tool and Materials Engineering, Faculty of Engineering, King Mongkut’s University of
Technology Thonburi, Thailand (ryan.mcc@kmutt.ac.th)
3
Center of Excellence on Petrochemical and Materials Technology, Chulalongkorn University,
Thailand (rojana.p@chula.ac.th)
Finite element analysis (FEA) has played a significant role in studying both the ballistic impact
process, as well as in optimizing the design of laminated transparent armor systems. Using FEA, the
goal of this research is to develop a lighter weight soda-lime glass/polyvinyl butyral (PVB) laminated
transparent armor, capable of NIJ level III ballistic protection. The projectile was 7.62x51mm NATO
ball having a maximum velocity of 838±15 m/s, as per the NIJ standard. The armor and projectile were
modeled in 3D using quarter symmetry. The FEA was performed using ANSYS explicit STR software.
The model was initially calibrated using experimental results obtained from a series of ballistic tests
against glass/PVB laminated armors. The effects of the glass (19, 15, and 12 mm) thicknesses, and the
glass order configuration, on the ballistic performance were systematically examined. The FEA results
were analyzed and compared to experimental ballistic results in terms of the damage in each glass layer,
the distribution of internal energy in the laminates, the bullet penetration depth, and the dimensions of
the impact crater. These results were subsequently used to evaluate the ballistic performance of
laminated armor designs and to identify the optimal design for lighter weight laminated armor as
compared to a reference design.
Keywords: Soda-lime glass laminates; Transparent armor; Finite element analysis; Ballistic protection;
Optimum design
113

1130 - 1150 hours
Finite Element Analysis of Dynamic Crushing of Functionally Graded

Foam-Filled Square Columns
Xuehui Yu, Qinghua Qin* and T. J. Wang
*
(qhqin@mail.xjtu.edu.cn)
This paper deals with the axial crushing and energy absorption of the functionally graded
foam-filled square columns subjected to axial impacts by using the finite element method. The
materials of square columns and foam used here are the same aluminum alloy. Effects of impact
velocity, wall thickness and the density distribution of graded foams upon the resulting specific energy
absorption are discussed in detail. It is shown that the functionally graded foam-filled square columns
are more attractive in energy absorption than uniform foam-filled columns when the impact velocity
attains a certain value.
Keywords: Functionally graded foam; Dynamic crushing; Energy absorption; Square column
114

1150 - 1210 hours
Structural Optimization of K4-Rating Removable Anti-Ram Bollards based

on Orthogonal Design Method
Suwen Chen1a, Tianyi Liu2, Guo-Qiang Li1b and Jian-Yun Sun3

1
State Key Laboratory for Disaster Reduction in Civil Engineering, China
a
(swchen@tongji.edu.cn), b(gqli@tongji.edu.cn)
2
College of Civil Engineering, Tongji University, China (liu_sky1@163.com)
3
China State Construction Engineering Corp. Ltd., China (sunjianyun96@163.com)
Anti-ram bollard systems, which are installed around buildings and infrastructures, can prevent the
entering of unauthorized vehicles, keep the distance from vehicle-borne improvised explosive devices
(VBIED) and reduce the corresponding damage. Compared with fixed bollard system, removable
bollard system provides more flexibility as it can be removed when needed. This paper proposed a new
type of K4-rating removable anti-ram bollard system. For simulating the collision of a vehicle hitting
the bollard system, a finite element model was built and verified through comparison of numerical
simulation results and existing experimental results. Based on orthogonal design method, the factors
influencing the safety and economy of this proposed system were examined and sorted according to the
importance. An optimized design scheme was then produced.
Keywords：Removable; Anti-ram bollards; Optimization; Orthogonal design method
115

1210 - 1230 hours
Prediction of Penetration Depth of Projectiles into Concrete Based on

Support Vector Machines
Qiang Pan, Xinkuan Zou and Jichun Zhang*
School of Civil Engineering, Southwest Jiaotong University, China

*
(pqiang1985@163.com)
At present, in concrete damage effect, prediction of penetration depth is of great guiding

significance to the design and construction of protection engineering. There are problems for the
traditional prediction methods, such as a high demand for samples, a big prediction error, and so on.
This paper applies a new intelligent algorithm, Support Vector Machines (SVMs), to build a model of
prediction of penetration depth. Parameters are optimized by using Particle Swarm Optimization (PSO)
and relative programs are written to predict penetration depth. And the predicted values are verified
according to the measured data of penetration depth. The results show that compared to the relative
error predicted by the traditional Grey Theory which is less than 9.29%, that predicted by SVMs which
is less than 3.18% is relatively smaller. This model has a great advantage for prediction of small
samples and nonlinearity. It is reasonable and feasible that Support Vector Machines are applied to
prediction of penetration depth of projectiles into concrete, which provides a new intelligent algorithm
for predicting penetration depth.
Keywords: Support vector machines; Concrete; Penetration depth; Prediction
116

1330 - 1350 hours
Finite Element Analysis of Long Cylindrical Tubes Subjected to Internal

Explosions
H.J. Li1, C.J. Shen1, G. Lu2* and Z.H. Wang3
1
2
Faculty of Science, Engineering and Technology, Swinburne University of Technology, Australia
3
Institute of Applied Mechanics and Biomedical Engineering, Taiyuan University of Technology, China
*
(glu@swin.edu.au)
Cylindrical metal tubes may be used as protective structures to contain potential explosives inside.
The response of empty and sandwich tubes under internal explosive loading was investigated in this
paper. In the beginning, an experimental work on the empty tubes under explosion was designed and
carried out to capture the fundamental deformation patterns. Then, numerical simulations, employing
ANSYS/LS-DYNA, are presented to broaden the study. Validation of the numerical analysis was made
by comparing the deformation patterns with the experiment. Consequently, a parametric study was
carried out to determine the key factor in the explosive process. With a constant failure strain, it is found
that the deformation of the tube is related to the mass of the TNT (M), the thickness (h) and stand-off
distance(R). The results are summarized in a diagram and got an empirical formula for the deformation
of the tube.
Keywords: Cylindrical tubes; Explosive loading; Structural response; Numerical simulation.
117

1350 - 1410 hours
Blast Mitigation By Stiffened Composite Plates for Vehicles under Mine

Blast
R. Lohani1, M. D. Goel2, S. Pradyumna3 and V. A. Matsagar1
1
Department of Civil Engineering, Indian Institute of Technology (IIT) Delhi, India
(matsagar@civil.iitd.ac.in)
2
Environmental Materials Division, CSIR - National Environmental Engineering Research Institute
(NEERI), India (md_goel@neeri.res.in)
3
Department of Applied Mechanics, Indian Institute of Technology (IIT) Delhi, India
(pradyum@am.iitd.ac.in)
It is important to maintain a fine balance between mobility, firepower, and protection while
designing a protection vehicle against mine induced blast. This type of design is primarily governed by
allowable weight, protection level, and cost. Composite plates prove to be a better solution under such
circumstances wherein different materials are used together to achieve the desired protection level.
Hence, in the present investigation, stiffened composite plates are studied as blast mitigation devices in
vehicles undermine blast. Stiffened composite plates with ceramic material sandwiched between two
steel plates are analyzed under buried mine blast using ABAQUS. The results show the effectiveness of
lighter ceramic material, i.e. silicon carbide as compared to heavier, tungsten carbide undermine blast,
which helps in achieving the desired weight and volume of the protection vehicle without sacrificing the
mobility under a given blast scenario. Analysis is also carried out to investigate the effect of employing
the simplified uniform rectangular pulse, Friedlander’s wave equation, and Jones-Wilkins-Lee (JWL)
equation for simulating the blast so that it can be easily used by designers. The plate is modeled
considering the strain rate effects by employing Johnson-Cook (J-C) model. For mine blast, air and soil
are defined as Eulerian domain and coupled Arbitrary Langrangian-Eulerian (ALE) model is used to
investigate the peak displacement of the composite plates to compare their effectiveness in blast
mitigation.
Keywords: Mine blast; Vehicle protection; Ceramic material; ABAQUS; ALE
118

1410 - 1430 hours
Finite Element Analysis of FRP-Strengthened Reinforced Concrete Panels

under Blast Loading
Xiaoshan Lin1a, Y.X. Zhang2 and Chi King Lee1b
1
a
(xslin@ntu.edu.sg), b(ccklee@ntu.edu.sg)
2
School of Engineering and Information Technology, the University of New South Wales, Australian
Defence Force Academy, Australia (y.zhang@adfa.edu.au)
A finite element model is developed in this paper for the simulation of structural responses of
FRP-strengthened reinforced concrete panels under blast loading. The strain rate effects for concrete in
tension and compression are considered separately in the current finite element model, and those for
steel reinforcements and FRP sheets are taken into account in their respective material models. The
effect of element size on the concrete material model under dynamic loading is also considered in the
numerical modelling. The commercial explicit hydrocode LS-DYNA is employed to carry out the
numerical analysis. A GFRP-strengthened reinforced concrete panel under blast loading is simulated in
this study, and the proposed finite element model is validated by comparing the computed results with
test data from the literature.
Keywords: Fibre reinforced polymer (FRP); Reinforced concrete panel; Finite element analysis; Blast
loading; Strain rate effect
119

1430 - 1450 hours
Effect of Blast Load on Response of Concrete Rectangular Storage Tanks

Primoz Jelusica and Bojan Zlenderb
Faculty of Civil Engineering, University of Maribor, Slovenia

a
(primoz.jelusic@um.si), b(bojan.zlender@um.si)
In this study, the time history response of concrete rectangular storage tanks subjected to blast load
is investigated. The importance of the modulus of subgrade reaction of the supporting soil is evaluated.
The effects of soil-structure interaction was considered in estimating the various responses. The
deflection and bending moments subjected to static and blast load was compared. It is concluded that
the modulus of subgrade reaction has larger influence on bending moment and deflection when
subjected to blast load in comparison with static load.
Keywords: Numerical; Blast load; Soil-structure interaction; Rectangular tank; Reinforced concrete
120

1450 - 1510 hours
Fast Analysis of Safety Areas for EOD in Urban Environment

Stefan Greulich1*, Thomas Hartmann1, Rene Richter2
1
NUMERICS, Germany
2
Ing.-Büro Richter, Germany
*
(stefan.greulich@numerics-gmbh.de)
Since World War II the disposal of unexploded ordnance has been a continuous task in Europe.
However, also in military missions outside Europe the explosive ordnance disposal (EOD) is
day-to-day business. Especially in urban environments, this is often an extremely challenging
undertaking, as the potential effects of a detonation in densely populated areas are usually significant.
Existing guidelines for explosive ordnance disposal are mostly based on regulations and expertise,
which are up to 70 years old and essentially based on German military requirements for steel
demolition. In these guidelines essential parameters such as the geometry, the position, the weight and
the type of the contained explosives will not be considered. Documents like the German military ZDV 3
/ 701 [1] usually demand large safety ranges (up to 1000 m) that are either hardly realizable or at least
very costly in modern urban environments. However, as detailed simulations of detonation events are
usually time consuming and the numerical simulation of the fragmentation process and fragment flight
is hardly feasible at all, the determination of safety ranges is often not backed by a physical analysis of
the specific scenario. This usually leads to a significant over-estimation.
In order to overcome this problem and to provide a physical foundation for the assessment of
safety ranges, in the following an approach to a rapid, physics-based estimation of safety ranges for
both blast and fragments generated by the detonation of munitions in an urban environment is
presented. As an application example, the recent blasting of a 250 kg bomb in the city of Munich is
analyzed. Fragment dispersion is assessed in a Monte Carlo analysis and blast propagation is simulated.
The resulting safety area is then compared to both the theoretical (according to guidelines) and the
actually evacuated zone.
Keywords: EOD in urban environments; Blast and fragments analysis; Numerical simulations
121

1330 - 1350 hours
Validation of Material Models for Design of Protective Structures

David Morin* and Magnus Langseth
Structural Impact Laboratory (SIMLab),Centre for Research-based Innovation and Department of

Structural Engineering, Norwegian University of Science and Technology, Norway
*
(david.morin@ntnu.no)
Due to the high cost of full-scale testing in fortification design, strong emphasis has to be placed on
the use of numerical simulations for load and structural response calculations. Such numerical tools can
thus be used for any design and risk assessment calculations as well as providing support to more
simplified design tools and design codes. Even though considerable advances have been made in
modelling, the designer must still use knowledge about the physical mechanisms controlling the
response. The designer must also know what simplifications can be made in the modelling and still
retain sufficient reliability and accuracy. At all levels of modelling, experimental validation of the
numerical models to be used is required before the models are accepted. The present work proposes a
systematic verification and validation procedure for material models used to design protective
structures with examples from the automotive industry. The approach is based on accurate experimental
tests as well as sound numerical modelling techniques.
Keywords: Impact; Material modelling; Validation
122

1350 - 1410 hours
On the Effectiveness of Blast Mitigation with Soft Resistance Claddings

Hongyuan Zhou1a, Zhiye Zhao1b and Xiaojuan Wang2
1
a
(hyzhou@ntu.edu.sg), b(czzhao@ntu.edu.sg)
2
Department of Civil and Environmental Engineering, National University of Singapore
(ceewang@nus.edu.sg)
Some buildings and infrastructures, potentially subjected to explosions either by accidents or

attacks, should be properly protected. In addition to the strengthening methods such as with fiber
reinforced polymer, an alternative is to apply sacrificial claddings with cellular solid core to the exterior
of existing buildings and structures. Some materials and structures with soft resistance and low relative
density will undergo large compressive strains with almost constant stress levels, absorbing large
amounts of energy. These are believed to be an ideal core for sacrificial claddings. Applying such
sacrificial claddings has multiple merits, including efficiency in protection, resilience, sustainability,
and flexibility. However, there are still some problems to be addressed before such claddings can be
successfully applied in engineering practice. In this paper, the underlying mechanism of blast mitigation
with soft resistance claddings is analyzed and the effectiveness is discussed. A criterion is established,
using energy approach, to determine the effectiveness of blast mitigation with soft resistant claddings.
Keywords: Blast mitigation; Soft resistance; Cladding
123

1410 - 1430 hours
Blast Performance of High Strength Concrete Panels: Experimental,

Numerical and Analytical Investigation
Satadru Das Adhikarya, Lado Riannevo Chandrab, Abraham Christianc and Khim Chye,
Gary Ongd

a
(ceesda@nus.edu.sg), b(ceelrc@nus.edu.sg), c(abraham@nus.edu.sg), d(ceeongkc@nus.edu.sg)
The Eurocodes will replace Singapore's prescribed building codes for structural design with effect
from 1 April 2015 when the current 2-year co-existence period ends with the prevailing
Singapore/British Standards being withdrawn. CP 65, based on the now obsolete BS8110, allows the
usage of concrete grade up to C50/60 whereas Eurocode permits concrete strength classes up to
C90/105. Since, such high strength concrete (HSC) mixes can be produced by most if not all the
ready-mix suppliers in Singapore, information pertaining to the use of HSC for protective structures
should also be available. The aim of the current research is to investigate the blast performance of HSC
panels and compares its performance with normal strength concrete (NSC) panels. Along with an blast
test program, 3-D nonlinear finite element (FE) modeling was also carried out for benchmarking.
Results from the simplified equivalent Single-Degree-of-Freedom (SDOF) model were also used. The
results obtained are presented and discussed in this paper.
Keywords: High strength concrete; Panels; Blast testing; Numerical simulation; Analytical model
124

1350 - 1410 hours
The Behaviour of Steel and Composite Beam-Column Joints under a

Middle-Column-Removal Scenario
Bo Yang1*,2, Kang Hai Tan3, Yongqing Chen1*,2
1
School of Civil Engineering, Chongqing University, China
*
(yang0206@cqu.edu.cn)
2
Key Laboratory of New Technology for Construction of Cities in Mountain Area (Chongqing
University), Ministry of Education, China
3
School of Civil & Environmental Engineering, Nanyang Technological University, Singapore
The objective of this research is to study the behaviour of steel and composite beam-column joints
under a middle-column-removal scenario. Firstly, the behaviour of different types of steel bolted
beam-column joints under a middle-column-removal scenario has been studied by experimental tests.
From the test results of the first series, it is found that bolted-angle connections normally performed
better than other types of connections in the development of catenary action. In order to achieve a better
understanding on the performance of bolted-angle connections under a middle-column-removal
scenario, nine experimental tests were carried out on different types of bolted-angle beam-column
joints. A new mechanical model has also been developed to predict the behaviour of bolted-angle
connections under pure tension. This new mechanical model is proposed based on a series of
component tests. In this model, the deformation capacities of bolted-angle connection components can
be predicted. Finally, a series of experimental tests was conducted in order to investigate the behaviour
of composite beam-column joints under a middle-column-removal scenario and identify the
contribution of composite slab to resist progressive collapse. The experimental tests consisted of two
types of tests, namely, middle joints under sagging moment and side joints under hogging moment. A
total of five composite beam-column joints with re-entrant steel profile decking were tested. The failure
modes, the contribution of composite slab to resist progressive collapse and the key components which
control the failure of the composite joints, were discussed in this study. In addition, a component-based
model was developed to predict the responses of composite beam-column joints subjected to a
middle-column-removal scenario.
Keywords: Progressive collapse; Steel and composite structures; Middle-column-removal scenarios;

Beam-column joints
125

1450 - 1510 hours
Damage Assessment of Reinforced Concrete Frames under Progressive

Collapse
Jian Weng*, Chi King Lee, Kang Hai Tan and Namyo Salim Lim

*
(jweng004@e.ntu.edu.sg)
It is well known that when a structure undergoes progressive collapse, severe damages propagate
under large deformations with the combined actions of flexural, shear and axial failure of structural
components. Therefore, a valid assessment scheme to identify crucial structural damages is essential for
progressive collapse analysis. In this paper, a damage assessment scheme was proposed to monitor the
growth of flexural, shear and axial damages for reinforced concrete (RC) frame structures. The
proposed assessment criteria considered the effects of axial-shear-flexural interactions during load
history. The respective flexural, shear and axial damages were quantified by appropriate damage indices
to trace the damage aggravation of structural elements. Finite element models adopting the suggested
damage assessment scheme were implemented and discussed. A numerical study was conducted and
compared with results obtained from the progressive collapse tests of three scaled RC frames.
Numerical and test results show that the proposed assessment scheme is capable of capturing critical
damage characteristics of the test frames.
Keywords: Progressive collapse analysis; Damage assessment; RC frames; Axial-shear-flexural

interactions
126

1540 - 1600 hours
Numerical Modeling of Projectile Explosive Charge Damage

during Penetration
Youjun Ning1*, Ge Kang2, Pengwan Chen2 and Xiaoping Zhang3
1
School of Manufacturing Science and Engineering, Southwest University of Science and Technology,
China
2
State Key Laboratory of Explosion Science and Technology, Beijing Institute of Technology, China
3
Institute of Geology and Geophysics, China Academy of Sciences, China
*
(cnningyj@foxmail.com)
In this paper, a numerical modeling technique based on the discontinuous deformation analysis
(DDA) method is used to simulate the dynamic damage process of explosive charge in a projectile
during penetration. A newly improved sub-block fracturing modeling method together with a newly
developed element elimination algorithm is used to simulate the damage of the charged explosive as
well as the failure of the concrete target. A modeling example is given. The penetration of a projectile
into the target and the dynamic response and damage process of the explosive charge are numerically
reproduced and analyzed.
Keywords: Projectile penetration; Charge failure; Numerical simulation; Discontinuous deformation

analysis (DDA)
127

1600 - 1620 hours
Energy Based Load-Impulse Diagrams

Ying-Kuan Tsaia and Theodor Krauthammerb
Center for Infrastructure Protection and Physical Security, University of Florida, USA
a
(tsai0406@ufl.edu), b(tedk@ufl.edu)
A load-impulse (P-I) diagram is a useful tool in preliminary strength design and damage
assessment. Current approaches to derive P-I diagrams are primarily based on numerical dynamic
analyses of the structural system. Further, there are no general energy based solutions that define P-I
relationships, and very little has been done to address energy flow as a basis for P-I diagrams. Since the
behavior of a structural member is governed by the amount and rate of external energy input and its
ability to dissipate energy, defining energy flow relationships will characterize the entire domain of
structural responses and corresponding damage. This paper presents a comprehensive energy flow
approach to enable one to define the entire P-I domain, and illustrates the application of an
energy-based P-I diagram. The results are validated using test data and numerical tools, such as DSAS
and ABAQUS/Explicit.
Keywords: Load-Impulse Diagrams; Energy Flow; SDOF; Impact; Blast
128

1620 - 1640 hours
Effect of Seismic Barrier on Response of Underground Structures Subjected

to Ground Shock
Jia Han Chew* and Eng Choon Leong

*
(jchew008@e.ntu.edu.sg), (cecleong@ntu.edu.sg)
Underground structures have many benefits compared with aboveground structures. They have
been used since early times as protective structures. To make underground structures more resilient
against ground shock, an additional protective layer has become a design option in recent times. The
protective layer serves as a barrier to lessen the transmission of the seismic waves rather than to protect
of the structure. Such seismic barriers usually come in the form of a low impedance layer such as air,
geofoam or cellular cladding. This paper investigates the use of a modified soil as the seismic barrier
against groundshock in underground structures numerically. Parameters such as peak pressure and peak
acceleration were used to examine the efficacy of the protective barrier. From the numerical
simulations, the stiffness of the barrier was found to have a direct impact on the magnitude of the
seismic loading on the structure. Softer soil as a seismic barrier tends to perform better in reducing the
seismic loading on the underground structure.
Keywords: Groundshock; Underground structure; Seismic barrier; Soil-rubber mixtures, Numerical

simulations
129

1640 - 1700 hours
Performance of Pile Groups Subjected to Surface Explosion

Laddu Bhagya Jayasinghe1*, Teck Chee, Anthony Goh1 and David Thambiratnam2
1
2
School of Civil Engineering & Built Environment, Queensland University of Technology, Australia
*
(ljayasinghe@ntu.edu.sg)
With the increased incidence of terrorist attacks over the world in the past few decades, the
importance of response of structures, including their foundation to credible blast loads has become a
vital factor in structural designs. Pile foundations are used in large weight structures to transfer the loads
from the superstructure above through weak compressible soil strata into deeper, stiffer soils which
have capacity to carry these loads. Piles are usually used in groups and the response of a pile within a
group could be different from that of a single pile because of the pile-soil-pile interactions that take
place in the group. This paper presents a numerical study on the response of pile groups to surface blasts
using Arbitrary Lagrangian-Eulerian technique.
Blast wave propagation in soil, horizontal pile deformation and pile damage are presented to
facilitate failure evaluation of piles. Effects of the number of piles and the spacing between piles within
a group on their blast response are investigated. The modeling technique developed and applied in this
paper and the findings provide a comprehensive understanding on the response of pile groups to ground
shocks induced by a surface explosion.
Keywords: Pile groups; Explosion; Damage; Soil; Numerical simulations
130

1700 - 1720 hours
Dynamic Response of Water Landing Capsule Body Based on The

Mesh-Free Method
Wang Yonghu1*, 2, Li Yulong1 and Chen Caihua2
1
2
Civil Aviation Flight University of China, China
*
(wangyonghucn@163.com)
A mesh-free method, Smoothed Particle Hydrodynamics (SPH), is performed to solve water

landing problem in order to carry out the crashworthiness analysis. Numerical simulation of space
capsule vertical water landing is implemented based on SPH method to verify the SPH-FEM model
reliability in LS-DYNA. The fluid is modeled with inner Smoothed Particle Hydrodynamics and lateral
Finite Element Method (FEM). The rigid structure employs FEM with Lagrangian formulation. The
optimal parameters, including particle density, contact stiffness, etc are studied by comprehensive
parametric analysis. The results obtained with the SPH-FEM coupled model show good agreement with
the given data in the paper and show that the SPH-FEM model is a viable and effective numerical tool.
Further studies were also conducted by a series of numerical simulations carried out to investigate the
load variation on the structure with various water-entry pitch angles. At last, the numerical results were
discussed and recommendations provided for future development of aircraft ditching and related water
landing problems.
Keywords: Water landing; Dynamic responses; SPH; Numerical simulations
131

1050 - 1110 hours
Modelling Dynamic Tensile Failure of Quasi-Brittle Materials Using

Stress-Enhanced Non Local Models
L. Pereira1*,2, J. Weerheijm1,3 and L.J. Sluys1
1
Faculty of Civil Eng. and Geosciences, Delft University of Technology, Netherlands
2
Portuguese Air Force Academy, Portugal
3
TNO, Defence, Safety and Security, Netherlands
*
(L.F.MagalhaesPereira@tudelft.nl)
The development of realistic numerical tools to efficiently model the response of concrete
structures subjected to close-in detonations and high velocity impacts has been one of the major quests
in defense research. Under these loading conditions, quasi-brittle materials undergo a multitude of
failure (damage) mechanisms. Dynamic tensile failure (e.g. spalling), characterized by a significant
strength increase associated with loading rate, has revealed to be particularly challenging to represent.
This phenomenon has been modeled by means of continuous damage mechanics in the last
decades. To minimize pathological mesh sensitivity, a nonlocal formulation is generally considered.
Nevertheless, these models fail to properly represent damage initiation and growth around
discontinuities, such as notches, damage areas and free boundaries. These inconsistencies are the
consequence of using a fixed interaction domain (characteristic length) in the nonlocal formulation. In
spite of limited experimental knowledge about the definition of the characteristic length parameter,
there is now consensus that this quantity is not constant.
In this contribution, an enhanced nonlocal model, where the interaction domain of any gauss point
contracts or expands according to the stress-state of its neighbors, is used. This formulation was coupled
to the well-known Mazars damage model and implemented within the framework of LS-DYNA using a
fully explicit computation scheme.
Two sets of numerical studies are presented in this paper. One shows the applicability and
limitations of the implemented explicit algorithm to compute nonlocal quantities. The other shows that
with this stress-enhanced regularization model it is possible to represent damage initiation and growth
more realistically and correct the inconsistencies emerging from the traditional nonlocal formulations.
Keywords: Concrete; Damage; Nonlocal; Erosion; Tensile failure
132

1110 - 1130 hours
Development of Modular Passive Attenuation Systems

Hui Wei Yeoa, Chor Boon Ngb, Ming Shu Tongc, Teng Sheng Pehd and Oi Yin, Karen
Chonge

a
(yhuiwei@dsta.gov.sg), b(nchorboo@dsta.gov.sg), c(tmingshu@dsta.gov.sg),
d
(ptengshe@dsta.gov.sg), e(coiyin@dsta.gov.sg)
A common challenge in designing protective facilities lies in the protection of ventilation openings
through which air is brought into and out of the facility. Smaller openings might be fitted with active
blast protection devices such as blast valves to reduce blast ingress. However, these solutions would not
be feasible for protecting larger openings. Modular passive attenuation systems were hence designed to
tackle the issue.
Modular passive attenuation systems are protective structures that lead air to flow into a maze-like
attenuation space comprising internal junctions (cross-junctions, U-junctions, T- junctions etc.). Blast
ingress into passive attenuation systems are expected to be attenuated to a degree that corresponds to the
complexity of the junction layout, and size of the attenuation space. A larger opening or larger threat
would call for increased complexity of the internal features and/or a larger attenuation space. Being
modular, such systems can be easily integrated into construction designs for protected facilities. The
paper on modular passive attenuation systems will describe the design considerations behind its
development, and introduce a sample design for illustration of the attenuation concept.
Keywords: Protective structure; Blast attenuation
133

1130 - 1150 hours
Assessment of Fast-Running Algorithms for Tunnel Blast Predictions
Teng Sheng Peha, Ya Dong Linb, Joel Ke Wei Ongc, Chor Boon Ngd and Karen Oi Yin
Chonge

a
(ptengshe@dsta.gov.sg), b(lyadong@dsta.gov.sg), c(okewei@dsta.gov.sg), d(nchorboo@dsta.gov.sg),
e
(coiyin@dsta.gov.sg)
The ability to evaluate blast loads in confined areas is necessary for the conduct of vulnerability
assessment against blast effects on critical systems in confined tunnel systems. These blast loads can be
predicted by computational blast modelling, or analytically based on a series of look-up charts. The
former has been shown to be accurate but is computationally expensive and requires modelling
expertise. For the latter, complex geometries that comprise many bends, expansions, constrictions and
junctions would require users to look up many charts. Besides being time consuming, the process is also
prone to human error. A fast-running tool was developed to automate the look up chart process and the
subsequent calculation procedures. Comprehensive studies were conducted to compare the results
derived from the fast-running tool, which are based on the charts, against those from computational
blast modelling and test data. In addition, some common geometrical configurations that were not found
in existing charts were developed and added to the fast-running tool to allow for a more comprehensive
analysis. The paper describes the development process of the fast-running tool, the validation work
done and some of its potential applications.
Keywords: Fast-running tool; Look-up charts; Blast effects; Confined passageways; Tunnel systems;
Computational blast modelling.
134

1130 - 1150 hours
Numerical Modelling of Chemical Dispersion via Elevated Air Intake
Wee Teik Khoo*, Kang Wei Lim, Shu Herng Ng, Heng Chye Chng and Yip Choy Lau
Ministry of Home Affairs, Singapore

*
(KHOO_Wee_Teik@mha.gov.sg)
Chemical agent attack is one of the possible terrorist attack scenarios considered by the
Ministry of Home Affairs. It is very costly and technically challenging to conduct full scale chemical
dispersal trials for consequence assessment purposes. With the advancement in computational
technologies, Computational Fluid Dynamics (CFD) can now be employed as a cost effective
alternative for the prediction of airflow, contaminant concentration and transportation within a building.
This paper examines the consequences of an outdoor chemical attack perpetrated via an elevated air
intake of a typical mechanical ventilated office building using numerical methods. A 3D, two-equation,
k-ɛ, turbulence model was developed using Commercial-off-the-shelves (COTs) CFD package, Fluent,
to study the spatial and temporal distribution and concentration level of the contaminant as it enter the
evaluated air intake. The technique used to solve for a transient condition, e.g. releasing a chemical
agent at a constant rate in a steady state environment, would also be discussed.
Keywords: Chemical attack; Air intake; Numerical simulation
135

1130 - 1150 hours
A Cellular Automaton Method for 2-Dimansional Solid Mechanics

Jishuo Wang and Weifeng Yuan*
China
*
(yuanweifeng@swust.edu.cn)
In this study, a numerical method that is used to the simulation of 2-dimensional elastostatics is
proposed based on the concept of cellular automaton (CA). By this method, an elastic domain is
discretized by a grid of nodes distributed randomly. To define a CA rule, it is assumed that the
displacement of a typical node is determined by the displacement of its neighbouring nodes located
within its influence zone covered by a square. The interpolation function for the conventional triangle
element used in finite element method (FEM) is employed to establish the relationship between the
typical node and its neighbours. The proposed method is totally a meshless method. The correctness of
the present approach is verified by numerical example.
Keywords: Cellular automaton; Heat conduction; Evacuation; Meshless
136

Protection Concepts and Design

PROTECTION CONCEPTS AND DESIGN ABSTRACTS

1330 - 1350 hours
Performance Based Framework for Resilient Design of Structures to Blast

Loads
Serdar Astarlioglua, Shalva Marjanishvilib and Francois Fayadc
Hinman Consulting Engineers, USA

a
(serdar@hce.com), b(shalva@hce.com), c(ffayad@hce.com)
This paper proposes a framework for quantifying resilience of large-scale buildings and bridges to
hazards such as blast and fire. To identify relevant scenarios that may lead to progressive collapse, the
damages that occur as a consequence of each hazard are established. The robustness of the structure is
considered as a function of the layout of the load-resisting elements of the structure and the relative
locations of the hazards. Resiliency is defined as a function of consequence of each threat and a
quantitative number for relative resiliency is established by inverting the consequence function. The
resiliency index can be used in the early stages of design the identify design decisions that will lead to a
more robust design. An example study that compares the relative resiliencies is provided.
Keywords: Progressive Collapse; Resiliency; Robustness; Performance Based; Air-blast
140

1350 - 1410 hours
Guidelines for Level of Protection of Facility
Dongfack Guepi Clovis Jamot1a, Sung-Wook Kim2 and Jong Yil Park1b
1
Department of Safety Engineering, Seoul National University of Science and Technology, Korea
a
(gueclovis@gmail.com), b(jip111@seoultech.ac.kr)
2
Structural Engineering Research Division, Korea Institute of Civil Engineering and Building Technology,
Korea (swkim@kict.re.kr)
There is an increasing need within many countries to protect people and facilities. But many of
them like the Republic of Korea have limited regulations or guidelines to handle this issue. So, LoP can
be considered as a starting point. The aim of this paper is to present a process to determine LoP and
allowable damage. LoP should be based on cost because, LoP is determined by asset value, which was
defined as "cost induced when an asset was damaged". Direct and indirect costs are considered to
include social and political damages and functional losses. For example, collateral damage outside of a
facility by hazmat leakage, and political sensitivities from mass causalities, can be considered in
determining asset value. Depending on LoP, the allowable level of structural damage is assigned. For
practical use, equations for each types of asset are suggested.
Keywords: Terrorism; Facilities; LOP; Asset value
141

1410 - 1430 hours
Risk Protec CI – Research Project for Aircraft Impact funded by the EU

Alexander Sieferta and Fritz-Otto Henkelb
Wölfel Beratende Ingenieure GmbH + Co.KG, Germany

a
(siefert@woelfel.de), b(henkel@woelfel.de)
As human-induced events and the related risk are important in the management and design
of critical infrastructures, the EU started the research call CIPS 2010 II. Within this call the project Risk
Protect CI under the leadership of Wölfel was realized [1]. Its goal was to improve an existing
numerical procedure using the finite element method (FEM) for the vulnerability analysis of critical
infrastructures as e.g. nuclear reactor buildings with respect to aircraft impacts.
In a first step the general procedure of the vulnerability analysis is described. Then in subsequence
the used material approaches, the aircraft models and the simulation method itself are validated in
comparison to available test data. Afterwards the improved method is applied for two exemplarily
designs of reactor buildings. Covering realistic scenarios the impacts of a B747 and a A320 with
varying velocities are investigated. The evaluation of the protective strength is carried out via the
observed damage of the outer containment.
Based on these results sensitivity studies with respect to the flight approach direction, the impact
point and the effect of obstacles are performed. Finally a procedure for improving the protective
strength is developed in terms of local measures. The concepts are investigated in detail and a benefit
for the protection capacity is determined.
The paper will show an overview of the project and look more detailed on the modelling approach,
the impact simulations, the sensitivity studies with respect to flight angle respectively impact point and
the benefit due to local measures to improve the design.
Keywords: Critical infrastructure; Aircraft impact; Numerical simulation; Protective design
142

1430 - 1450 hours
Quantitative Risk Analysis of Gas Explosions in Tunnels; Probability,

Effects, and Consequences
J. Weerheijm1,2, M.M. van der Voort1*, J. Verreault1, A.C. van den Berg1
1
TNO, Netherlands
2
University of Technology Delft, Faculty of Civil Engineering, Netherlands
*
(martijn.vandervoort@tno.nl)
Tunnel accidents with transports of combustible liquefied gases may lead to explosions.
Depending on the substance involved this can be a Boiling Liquid Expanding Vapour Explosion
(BLEVE), a Gas Expansion Explosion (GEE) or a gas explosion. Quantification of the risk of these
scenarios is important to take informed decisions on tunnel design and routing of dangerous goods. In a
risk analysis scenarios have to be taken into account each with their own physical effects (overpressure
and fire), consequences (damage to the tunnel, lethality, injury), and probability of occurrence.
The current paper focuses on the gas explosion scenarios. First the gas explosion mechanism is
summarized to introduce the key parameters for the explosion load in a tunnel. Then an integrated
model for the dispersion and probability of ignition is presented and applied. Case studies with an
instantaneous LPG release show that the cloud is initially too fuel-rich to be ignited. Subsequently the
gas concentration either remains mostly fuel-rich by the time the cloud reaches the end of the tunnel or
falls within the flammability limits before the tunnel exit, depending on the initial amount of fuel
released. A case study with a continuous LPG release shows that depending on the release rate and the
ventilation speed, the tunnel may be completely filled with a combustible fuel-air mixture. The
simulations provide the overpressure for different scenarios linked with their probability of occurrence.
Keywords: Explosion; Tunnel design; Probability; Risks; Models; Transport dangerous goods
143

1450 - 1510 hours
An Engineering Approach for Risk, Resilience and Vulnerability

Assessment of Urban Areas
Kai Fischer*, Arno Klomfass and Werner Riedel
Fraunhofer Institute for High-Speed Dynamics, Ernst-Mach-Institute, Germany

*
(kai.fischer@emi.fraunhofer.de)
A large part of the world’s population, economic activity, and physical infrastructures is
concentrated in fast growing urban environments. Among these, the security and citizen safety in
densely populated areas has become a major issue. In view of the growing sensitivity to terrorism, large
scale accident scenarios, natural disasters and crime, urban planning practice must incorporate
appropriate security measures for vulnerability identification and resilience enhancements. In this
paper, a systematic approach is presented to evaluate possible threats and their expected consequences.
The software assisted procedure is based on suitably defined risk analysis and management schemes
and uses validated engineering tools for quantification. The paper provides a general description of the
approach and the underlying methods. As an example application the evaluation of hazards from
explosive threats is presented. The holistic approach is applied to assess decision variables like human
injury, structural or monetary damage. Based on empirical data the susceptibility of a city is calculated
to derive hot spots at risk. Afterwards a quantitative vulnerability approach including physical and
engineering models is applied to identify weak spots in an urban surrounding. The results deliver
decision support to see where countermeasures help most. For further detailed evaluation, a fast running
CFD tool for explosions has been seamlessly incorporated into the software and offers more precise
information concerning complex blast wave propagation. The results build the basis to derive security
measures for the generation of robust and sustainable cities.
Keywords: Urban planning; Explosive threats; Vulnerability; Resilience enhancement
144

1540 - 1600 hours
Analysis of Explosion Incidents in Mass Transport Systems

G Meyer1*, M Kumm1 & B Janzon2
1
School of Business, Society and Engineering, Mälardalen University, Sweden
*
(gero.meyer@mdh.se), (maria.kumm@mdh.se)
2
SECRAB Security Research, Sweden (bo.janzon@secrab.eu)
Underground mass transport systems are necessary for large cities in order to relax the traffic
situation on the surface. Millions of passengers rely on a functional system and gather on subways and
commuter trains to travel throughout the cities. Large numbers of people in small areas create
interesting targets for terrorists who intend to cause fear among the population worldwide. Recent
decades have shown an increased activity of antagonistic attacks involving explosive devices in rail
bound mass transport systems. A summary of the most serious attacks will offer an opportunity to learn
about consequences to passengers and employees and the response of building structures and rolling
stock. The paper compiles 19 incidents for direct comparison and presents numbers of casualties and
injured passengers. The main cause of injuries, the response of building structures and components as
well as technical details about type, size and location of the explosive devices are discussed. This paper
further provides some information on evacuation issues and the response by the fire and rescue services.
Problems faced by first responders are described and measures to facilitate the fire and rescue
operations are presented. The results were used as a basis for design of full- and model-scale commuter
train explosion tests carried out within the METRO project. The analysis shows that antagonistic
attacks are repeating irregularly in time and place and that openly available information is scarce.
Keywords: Literature review; Antagonistic attacks; Deliberate attacks; Terrorist attacks
145

1600 - 1620 hours
A Structured Approach to Forensic Study of Explosions: The TNO Inverse

Explosion Analysis Tool
M.M. van der Voort*, R.M.M. van Wees, S.D. Brouwer, M.J. van der Jagt-Deutekom and
J. Verreault
TNO, The Netherlands

*
(martijn.vandervoort@tno.nl)
Forensic analysis of explosions consists of determining the point of origin, the explosive
substance involved, and the charge mass. Within the EU FP7 project Hyperion, TNO developed the
Inverse Explosion Analysis (TNO-IEA) tool to estimate the charge mass and point of origin based on
observed damage around an explosion. In this paper inverse models are presented based on the two
most frequently occurring and reliable sources of information: window breakage and building damage.
The models have been verified by applying them to the Enschede firework disaster and the Khobar
tower attack. Furthermore a statistical method has been developed to combine the various types of data,
in order to determine an overall charge mass distribution.
In relatively open environments, like for the Enschede firework disaster, the models generate
realistic charge masses that are consistent with values found in forensic literature. The confidence
interval predicted by the IEA tool is however larger than presented in the literature for these specific
cases. This is realistic due to the large inherent uncertainties in a generic tool. Furthermore, to our
judgment often a too narrow range of charge masses is reported compared to the evidence. The
IEA-models give a reasonable first order estimate of the charge mass in a densely built urban
environment, such as for the Khobar tower attack. Due to blast shielding effects which are not taken into
account in the IEA tool, this is usually an under prediction. To obtain more accurate predictions, the
application of Computational Fluid Dynamics (CFD) simulations is advised.
The TNO IEA tool gives unique possibilities to inversely calculate the TNT equivalent charge
mass based on a large variety of explosion effects and observations. The IEA tool enables forensic
analysts, also those who are not experts on explosion effects, to perform an analysis with a largely
reduced effort.
Keywords: Explosion; Forensic analysis; Blast; Debris; Damage; Statistical analysis
146

1620 - 1640 hours
Buildings as Barricades Model (BBM)

Michael M. Swisdak, Jr.a and Tyler J. Rossb
APT Research, Inc, USA

a
(mswisdak@apt-research.com), b(tross@apt-research.com)
Buildings as Barricades Model (BBM) is a Microsoft Excel spreadsheet designed to estimate the
effects of buildings acting as barricades providing shielding effects from debris and overpressure to
other structures in the shadow of the initial building. The tool is intended to function as a quick
screening tool for hazards. The tool creates a simple plot showing pressure, impulse and debris contours
for an explosion of one or more ISO containers in a single location. Buildings can be defined on a two
dimensional Cartesian coordinate system to interact with the blast effects and cause shielding. The
shielding is based on only the physical dimensions of the building acting as a barricade and does not
account for actual structural response of the building. The strengths of the tool include the fast-running
nature and simplistic input of an Excel spreadsheet as well as the fast visual representation of the
results. Debris density as well as pressure and impulse can be reported for any point on the map. These
values are calculated using the advanced debris density fast running models developed for the Institute
of Makers of Explosives’ Safety Analysis for Risk (IMESAFR) software and airblast calculations from
the Department of Defense Explosives Safety Board (DDESB) Blast Effects Computer (BEC). The
model also accounts for and reports blast wave attenuation due to the shielding of the barricade
structure. The paper presents the technical assumptions and simplifications associated with creating this
screening tool. General software requirements are addressed and features of the tool are described in
detail. Finally, an example is provided that utilizes the various features of the tool.
Keywords: Barricade; Blast shielding; Debris; Fast running model
147

1640 - 1700 hours
Reducing Land Usage for Blast Protected JTC Standardised 66/22kV

Substations
Peter J Hoad1, Su Chern Tan2a and Chin Wai Aw2b
1
Arup Singapore Pte Ltd, Singapore (peter.hoad@arup.com),
2
JT Corporation, Singapore
a
(tan_su_chern@jtc.gov.sg), b(aw_chin_wai@jtc.gov.sg)
In 2013, JT Corporation (JTC) led a study brief to look at standardising the 66/22kV electrical
substation design in JTC Industrial Estates. The primary intents of the study included:
 Improving the construction efficiency.
 Pre-approval of Nation Critical Infrastructure (NCI) standardised substation designs with
Singapore Authorities.
 Optimisation of the building footprints to fit triangular and rectangular land parcels.
Arup was engaged by the JTC appointed Architect (LT&T) to advise on all blast resilience and other
security related matters.
In land scarce Singapore, achieving a desirable 25 metres stand-off distance to achieve economical
and blast resilient built form infrastructure would be difficult. With stand-off as little as 5 metres, the
following design measures were introduced to improve business continuity reliability associated with
an off-site VBIED threat:
 Reconfiguration of the (standardised) substation layouts to reduce blast related risks to the
critical substation infrastructure, and
 Introduce a site perimeter blast wall that acts to limit the pressures and impulses felt by the
substation building for a number of different VBIED detonation scenarios. This
methodology affectively allowed reduction of the stand-off distance from 15 metres
(minimum recommended for the design) to as little as 5 metres. This not only significantly
reduces the land take, but also allows the design to be used for a significant increase in plot
size variation.
Keywords: Blast wall; Spall; Breach; Scabbing
148

1700 - 1720 hours
A Novel Device for Blast Energy Dissipation

Hongyuan Zhou1a, Zhiye Zhao1b and Xiaojuan Wang2
1
a
(hyzhou@ntu.edu.sg), b(czzhao@ntu.edu.sg)
2
Department of Civil and Environmental Engineering, National University of Singapore
(ceewang@nus.edu.sg)
For decades, a number of structures and devices have been proposed to dissipate blast energy and mitigate
blast effects on buildings. These included for example, cellular solid/structure claddings. The underlying
mechanism in such systems involves the dissipation of considerable amounts of energy when the cellular
solid/structure claddings undergo large plastic deformation at almost constant stress, leading to controllable low
transmissions of load to the protected buildings/structures.
By contrast, a novel device is proposed in this study to dissipate blast energy; it does so by converting
translational energy to rotational energy, in a way which is not detrimental to the protected structure. In this paper
the energy conversion mechanisms are analyzed at different phases of the device’s functions, and the key
parameters governing energy conversion efficiency are identified and discussed. In the process, other possible
energy dissipation methods for strong, short duration loads are also identified.
Keywords: Blast mitigation; Energy conversion; Energy dissipation; Cladding
149

1540 - 1600 hours
Design, Analysis and Testing of Blast-Resistant Curtain Wall

Kenneth Morrill1, Shengrui Lan1* and Russell Norris2
1
Karagozian& Case, Inc., United States
2
U.S. Department of State Bureau of Diplomatic Security, United States
*
(lan@kcse.com)
This paper presents the design considerations for enhancing blast resistant capability from a
conventional curtain wall design, the HFPB simulation techniques and pretest calculations that were
used for determining the design parameters, and the full-scale blast test results of the curtain wall test
article. The blast test has proven the blast resistant design and high-fidelity physics-based (HFPB)
simulation results were in good agreement with the test results. The comparison between the HFPB
pretest predictions and the blast test results will be discussed. With the successful blast test validation,
the developed blast resistant curtain design can be used as a best practice technique for re-cladding
existing building envelops as well as new constructions. The project also demonstrated that the HFPB
simulation technique plays a very important role in development of such kind of systems because it
takes into account many complicated dynamic phenomena, including “real-time” distribution of blast
loading, large dynamic deformations, nonlinear material properties with stain rate effects, dynamic
interaction among the components, connection details, actual support conditions.
Keywords: Curtain wall; Blast resistant; Blast test; HFPB simulation; Façade
150

1600 - 1620 hours
The Use of Fiber Composites for Blast and Fragmentation Protection of

Selected Civilian Infrastructure
Moshe Ravid1, Nimi Shapira1, Zvi Asaf 2, Tamir Blass2, Barry L. Bingham3 and
Raymond H. Bennett3
1
Rimat Advanced Technologies Ltd., Israel (rimatoffice@rimat-ltd.com)
2
Plasan Sasa, Israel
3
Baker Engineering and Risk Consultants, Inc., United States (BBingham@BakerRisk.com)
Infrastructure applications such as refineries and petrochemical plants, power plants or off shore
Vessels, often require blast and fire resistance, and, in some cases, fragmentation protection as well.
Contemporary threats to be taken into account involve fire or gas explosion which may accidently
happened as well as attacks by blast and fragmentation warheads integrated in several types of rockets,
mortars and fragmentation generated by artillery HE blast shells.
Protection add–on for predesigned blast resistant solutions are needed vastly for crew
survivability. The main limitation for such solutions is their weight, especially when such solutions are
supported by the existing structure.
In this paper the authors will demonstrate the use of fiber composite add-on protective curtains, as
well as alternative light weight wall construction, designed to certain protection requirements, including
numerical simulation and live field tests evaluation tools, and discuss some practical conclusions.
Keywords: Impact; Impulsive; Shock; Protective design; Numerical simulations
151

1620 - 1640 hours
Nested Tube System Applicable to Protective Structures Against Blast

Shock
Ze Liang Yu, Pu Xue* and Z. Chen

*
(p.xue@nwpu.edu.cn)
Tubes and rings have been widely used as efficient energy absorbing components, due to their light
weight, providing large and stable deformation, and low cost. Many studies have been conducted on
their dynamic behavior and energy absorbing performance under lateral loading. However, single ring
or tube usually is not enough in energy absorption capacity because of its limited load-carrying
capacity. Therefore, some attempts have been carried out by using rings or tubes arranged in series or
parallel. Aiming at developing efficient energy absorbing components for protective structures against
blast shock, this paper proposes a nested tube system consisting of three tubes (TT) and its dynamic
performance is compared with that of a double-tube (DT) system and a single-tube (ST) system under
quasi-static and dynamic loading. The results show that TT system can provide the highest energy
absorption efficiency and the most stable deformation history, as well as and the lowest peak reaction
force, so as to reduce the damages for protected structures. Then, the deformation mechanisms of the
nested tube system, as well as their dependence on the thickness and diameter of tubes are extensively
investigated. Finally, the nested-tube (TT) system is applied to a blast-resistance door used in civil air
defense headquarters, showing its effectiveness in protection from damage under blastshock waves.
Keywords: Nested-tube system; Energy absorption; Protective structures; Blast shock
152

1640 - 1700 hours
Improving Robustness of Medium-Rise Composite Building Using Multiple

Belt-Truss Systems
C. G. Taya, Chan Ghee Kohb and J. Y. Richard Liewc

a
(cgtay@nus.edu.sg), b(cgkoh@nus.edu.sg), c(ceeljy@nus.edu.sg)
Previous studies by the authors have shown that belt-truss (BT) system can be an effective
robustness enhancement of multi-story composite building in the event of sudden column loss. This
paper extends the concept to explore the potential of BT system for medium-rise composite buildings,
with focus on the influence of the BT position and the use of multiple BT systems on its robustness
characteristics, namely the ductility and force demands. Our findings show that BT system is effective
in reducing ductility demand. However, its position places an important role and has to be carefully
considered to control the column force demands so as to avoid overstressing of the columns.
Alternatively, multiple BT systems can be adopted for better robustness performance albeit at a higher
cost. For robustness enhancement existing buildings, it is more practical to use single BT system,
whereas for the design of new buildings, multiple BT systems would provide a greater level of
robustness assurance.
Keywords: Structural robustness; Progressive collapse analysis; Sudden column loss; Belt-truss system;
Composite building.
153

1700 - 1720 hours
Performance of an Explosive Water-Mist Spray System for Dust Control in

Explosive Demolition
Hyon-Soo Kim1, Young-Hun Ko2, Hoon Park3, Chul-Gi Suk3, Hyung-Sik Yang4 and
Sang-Ho Cho5
1
Department of Mineral Resources & Energy Engineering, Chonbuk National University, Korea
(khs1221@jbnu.ac.kr),
2
Department of Energy Resources Engineering, Chonnam National University, Korea
(dddd8401@nate.com),
3
Korea Kacoh Co., Ltd., Korea (hujin@jbnu.ac.kr), (kacoh@hanmail.net),
4
Department of Energy Resources Engineering, Chonnam National University, Korea
(hsyang@jnu.ac.kr)
5
Department of Mineral Resources & Energy Engineering, Chonbuk National University, Korea
(chosh@jbnu.ac.kr)
Dust from explosive demolition should impact greatly on the health and quality of life of people
working on and living close to these sites. For the purpose of dust control in explosive demolition sites,
a water-mist spray system which uses explosion pressure of a detonating cord in a water tube was
suggested. In this study, the behavior of water-mist by produced by explosion in a P.E. Tarpaulin tube
containing water was analyzed and the mechanism of the dust reduction of water-mist was discussed. In
order to observe the ejection behavior of water-mist produced by explosion in a water tube, high speed
camera images were captured during the explosion. Using the analysis in the AUTODYN software, the
water-mist generation by explosion was simulated and the diffusion mechanism of water was discussed.
Furthermore, the Weber number, which defined as the ratio of the inertial force of the surface tension,
was evaluated with the expansion velocity and pressure of the cavitation and gas bubble generation in
the water tube.
Keywords: Explosive demolition; Water-mist; AUTODYN; Weber number; Gas bubble
154
Rock Dynamics
ROCK DYNAMICS ABSTRACTS

1330 - 1350 hours
Dynamic Properties of Water Saturated Sandstone

Zilong Zhou1*, Xin Cai1, Kun Du2, Cheng Xiong1, Yuan Zhao1
1
School of Resources and Safety Engineering, Central South University, China
2
Advanced Research Center, Central South University, China
*
(zlzhou@csu.edu.cn)
Water content plays an important role in affecting the mechanical properties of rock materials.
Water injection is even used for the prevention of some dynamic hazard like rock-burst. To investigate
the dynamic behavior of water-saturated rock under impact, Split Hopkinson Pressure Bar (SHPB)
technique was used to do compressive and tensile tests on sandstone specimens with different degrees
of saturation. The results show that both the dynamic compressive and the tensile strength decrease with
the increasing degree of water saturation. When the rock is well saturated, it tends to have more
plasticity characteristics than the dry rock.
Keywords: Impact; SHPB; Rock property; Water saturation
158

1350 - 1410 hours
Numerical study of dynamic fluid flow in rock fracture network using the
Lattice Boltzmann Method
Peijie Yin and Gao-Feng Zhao*
Centre for Infrastructure Engineering and Safety, School of Civil and Environmental Engineering,
The University of New South Wales, Australia
*
(gaofeng.zhao@unsw.edu.au)
The dynamic behavior of fluid flow in a rock fracture network is investigated using the Lattice
Boltzmann Method (LBM). The LBM has advantages on high efficiency for parallel computation,
ability of dealing with complex geometry and boundary conditions, and no need for the isolated and
dead-end detection in the fracture network. It was first verified against the analytical solution of the
corresponding pipe network model on modeling fluid flow in the rock fracture network. Then, the
influence of the length-density parameter on the dynamic fluid flow behavior was studied numerically.
It was found that the hydraulic conductivity of the rock fracture network increases with the
length-density parameter. Moreover, the dynamic fluid flow behavior associated with the uniaxial
deformation of the rock fracture network was studied through coupling the discrete element model
(DEM) and the LBM. The anisotropic behavior of the fluid flow behavior under uni-axial deformation
was successfully captured. Finally, the influence of different pressure gradients on the fluid flow in the
rock fracture network was investigated. The numerical results indicate that an apparent dynamic
behavior exists in the fluid flow through the rock fracture network, i.e. the hydraulic conductivity
cannot be treated as a constant under large pressure gradient.
Keywords: Rock fracture network; Dynamics; Lattice Boltzmann Method; hydraulic conductivity;
hydro-mechanical coupling
159

1410 - 1430 hours
Dynamic Ring Test of Rock under Diametral Impact Load by SHPB

Diyuan Li*, Tengjiao Cheng, Xibing Li and Junyu Qian
School of Resources and Safety Engineering, Central South University, China

*
(diyuan.li@csu.edu.cn)
Dynamic ring tests have been carried out on marble specimens with different internal diameters (d)
by a splitting Hopkinson pressure bar (SHPB) apparatus. Seven groups (A, B, C, D, E, F and G) of ring
marble specimens with a nominal external diameter (D) of 50 mm and a nominal thickness (t) of 25 mm
have been tested under impact load, including internal diameter ratios (r = d/D) ranging between 0 and
0.6 with an interval of 0.1. The incident stress waves, the reflected stress waves and the transmitted
stress waves are obtained by the strain gauges pasted on the incident and transmitted bars. Testing
results show that with the increase of the internal diameter ratios, the calculated dynamic tensile
strength of the marble decreases. The dynamic tensile strength of ring specimens is about 4.3 to 1.3
times more than that obtained by the dynamic Brazilian tensile test when r ranges between 0.1 and 0.6.
It indicates that the dynamic tensile strength demonstrated in ring test may be an experimental property
rather than a material property. During the impact ring test, a high speed camera is used to capture the
fracturing process of the ring disc marble specimens. It can be clearly observed that the diametral tensile
crack initiates from the center of internal hole along the compressive line to the loading point.
Afterwards, the bending tensile fractures initiate from the external of the ring to the internal hole almost
perpendicular to the compressive diametral line. Ring specimens usually fail with four main fan-shaped
rings. With the increase of the internal hole diameter, the bending tensile cracks will initiate near the
incident bar. It indicates that the stress balance may be difficult to achieve when the internal diameter
ratios increase to more than 0.4. In such a condition, the dynamic response of the ring specimen under
impact load will show more structure effect instead of material effect.
Keywords: Impact; Dynamic tensile strength; Ring test; Brazilian test; SHPB
160

1430 - 1450 hours
Numerical Simulation on Failure Process of Unloaded Rock Triggered by

Dynamic Disturbance
Leilei Niu, Wancheng Zhu*, Zhen Cheng, Jiong Wei and Chenhui Wei
Key Laboratory of Ministry of Education on Safe Mining of Deep Metal Mines, Northeastern
University, China
*
(zhuwancheng@mail.neu.edu.cn)
During the excavation of underground opening, the rock may experience a complex loading path
that includes the highly confined compression, unloading of confining stress and disturbance of
dynamic loading. By using a numerical code named Rock Failure Process Analysis for Dynamics
(RFPA-Dynamics), the failure of rock sequentially subjected to highly confined compression,
unloading of confining stress and disturbance of dynamic loading is numerically simulated in order to
examine the failure mechanism of rock induced by excavation. The RFPA-Dynamics is firstly used to
reproduce the failure of rock under confined compression and under unloading of confining pressure,
and it is validated against with the existing experimental observation. The numerical results indicate
that the arc-shaped damage zone is prone to develop and it is influenced by the vertical in-situ stress,
lateral pressure coefficient and both the amplitude and duration of the dynamic disturbance. This study
shows that it is of great significance to trace the complex loading path induced by excavation in order to
capture the rock failure mechanism induced by underground excavation.
Key words: Rock failure; Unloading of confining stress; Dynamic disturbance; Numerical simulation
161

1430 - 1450 hours
Blast Wave Propagation across Jointed Rock Masses and its Influence on
the Ground Motion
Shaobo Chaia, JianChun Lib and Haibo Lic
State Key Laboratory of Geomechanics and Geotechnical Engineering, Institute of Rock and Soil
Mechanics, Chinese Academy of Sciences, China
a
(553890073@qq.com), b(jcli@whrsm.ac.cn), c(hbli@whrsm.ac.cn)
Studying blast-induced wave propagation across rock masses and the induced ground motion is an
important topic which has been received considerable attention in the design and construction of
underground cavern/tunnel constructions and mining activities. The present investigation is concerned
with the interaction between the blast wave and a rock joint by using a time-domain recursive method.
Based on the conservation of momentum at wave fronts, two types of interaction are analyzed first, i.e.
the interaction between blast wave (i.e. longitudinal wave) and a linearly elastic joint and the interaction
between stress waves (longitudinal wave or transverse wave) and a free surface. Next, analysis for the
transmission and reflection processes for stress wave propagation between a single joint and the ground
surface is carried out. The wave propagation equation based on the proposed recursive method in time
domain for every wave arrival at the free surface is established when the displacement discontinuity
method and the time shifting function are considered. On the basis of the derived wave propagation
equations, the mathematical expressions are deduced for ground surface vibration caused by stress
waves. As the result of the wave superposition, the particle velocity on the ground surface is
subsequently obtained.
Keywords: Recursive analysis; Time domain; Blast wave; Wave propagation; Ground motion; Peak
particle velocity
162
Structural Behaviour and Damage

STRUCTURL BEHAVIOUR AND DAMAGE ABSTRACTS

1540 - 1600 hours
Joint Slippage Effects on the Impact Dynamic Response of Lattice Angle

Steel Tower Subject to Rupture of Conductor
Jiang Wenqiang* and Yang Wengang
Department of Mechanical Engineering, North China Electric Power University, China

*
(jwq1980@hotmail.com)
Transmission towers are vital components of overhead transmission lines. Wire rupture is an
important design condition in the prototype design of lattice transmission towers, which have great
impacts once it happens. Sudden wire rupture will not only cause transmission line outage. In some
severe cases it may also lead to whole transmission line collapse from domino effects. Traditionally,
lattice towers are analyzed without considering joint slippage effects. As a result great differences are
common, when analyses from idealizations are compared with full scale test results. In this paper two
numerical models of a latticed angle steel tower are given which include joint slippage effects, and the
equivalent model of transmission lines are created considering conductor and ground wire supporting
effects. The dynamic response of the lattice tower subject to the rupture of middle and right conductors
are analyzed. The influence of joint slippage effects on dynamic response and impact coefficient are
presented.
Keywords: Joint Slippage; Lattice Angle Steel Tower; Impact Dynamic Response; Rupture of
Conductor
166

1600 - 1620 hours
Local Effects Due to the Opening Sizes on the Negative Bending Regions of
Steel-Concrete Composite Beam Owing to Axial Compression
Ma. Bavan*, Shahrizan Bin Baharom and Siti Aminah Osman
Dept. of Civil and Structural Engineering, National University of Malaysia, Malaysia

*
(mmbavan@yahoo.com)
Existing research and design methods for web openings of steel concrete composite beam differ
widely in concept when it was subjected to combined negative bending and axial compression such that a
hypothesis had identified that probable unfavourable failure phenomenon causes with opening sizes. This
paper presents the local effects due to the opening sizes influence in terms of the developed stresses on
steel beam. A validated Finite Element (FE) model with experimental analysis from previous research
studies has been taken to account for investigating this concept such that openings with various sizes have
been introduced on steel beam and the failure modes in the material components have been analysed
throughout the simulation of the FE models developed. The results predicted were significant in terms of
the local and global effects. Further, it was examined the ultimate and local effects with introducing flange
and web stiffeners to improve the limit states by postponing the failure criteria.
Keywords: Composite beam with web openings subjected to combined negative bending and axial
compression; opening sizes, non-linear 3D FE model; local effects, postponing the failure criteria by
stiffeners
167

1600 - 1620 hours
An Efficient Analytical Failure Approach in the Global Analysis of a Buried

Pipeline System
Kwong Ming Tse1*, Long Bin Tan1, Heow Pueh Lee1, Beng Chye, Vincent Tan1, Ke Wei,
Joel Ong2, Chor Boon Ng2, Oi Yin, Karen Chong2
1
Department of Mechanical Engineering, National University of Singapore
2
*
(tsekm@nus.edu.sg)
In recent years, light-weight composite pipes have been used extensively in various industries.
However, their material anisotropy and discontinuity make both stress and failure analyses of such
multilayered structures complicated and cumbersome. An efficient analytical failure analysis approach
using an executable code based on a layer-by-layer progressive damage model and a homogenization
stress analysis method has been developed and is proposed for the analysis of glass-reinforced epoxy
(GRE) composite pipes used for fluid transportation. Initial and final failure envelopes in both stress
and force spaces are generated by the proposed approach. In addition, a spreadsheet based program has
also been developed to facilitate model definition for global analysis of any arbitrary pipeline system
under any given loading conditions. Two cases of an arbitrary pipeline system under both service road
and valve pit loadings, with and without pipe internal pressure, are presented in the study. It is found
that pipe bends are most vulnerable to axial tension and internal pressure during operation as significant
amount of axial force would be generated by the internal flow. By transforming the failure envelopes
into the bending moment-tension-pressure (MTP) space, these MTP values within the pipe network are
plotted directly onto the envelopes to assess critical locations of the composite pipe. This failure
analysis approach provides an efficient way for the design of composite pipeline system for fluid
transportation. However, precautions should be taken due to simplifications in the global model. For
example, joints and pipe-pit penetrations are not modelled in this global analysis.
Keywords: Pipe; Composite; Failure envelope
168

1050 - 1110 hours
Structural Behaviour of Precast Beam-Column Sub-Assemblages under

Shao-Bo Kang1a, Kang Hai Tan1b, En-Hua Yang1c and Kian Wee Ng2
1
School of Civil & Environmental Engineering, Nanyang Technological University, Singapore
a
(skang2@e.ntu.edu.sg), b(CKHTAN@ntu.edu.sg), c(EHYANG@ntu.edu.sg)
2
Civil and Structural Engineering Department, JTC Corporation, Singapore
(NG_Kian_Wee@jtc.gov.sg)
This paper presents an experimental study on the structural behaviour of precast beam-column
sub-assemblages under column removal scenarios, in which conventional concrete and engineered
cementitious composites (ECC) were used in the cast-in-situ concrete topping and the beam-column
joint. The specimens were restrained by horizontal and vertical load cells and tested under quasi-static
loading condition. Experimental results indicate that at the initial stage ECC and steel reinforcement
sustained tensile stresses compatibly and multiple cracking was observed in the ECC topping. When the
tensile strain capacity of ECC was exhausted, major cracks were formed in the plastic hinge region,
whereas at other sections ECC remained intact to transfer tensile stresses across cracks. Compared to
conventional concrete, ECC resulted in more severe localisation of rotations in a limited region due to
its tension-stiffening behaviour. Conclusions are drawn that application of ECC to structural topping
and beam-column joints did not significantly enhance structural resistance of sub-assemblages under
column removal scenarios. However, the calculated pseudo-static resistance of ECC sub-assemblage is
greater than that of concrete specimen due to better energy-absorption capacity of ECC.
Keywords: Column removal scenarios; Beam-column sub-assemblages; Engineered cementitious

composites; Compressive arch action; Catenary action
169

1110 - 1130 hours
Progressive Collapse Performances of Three-Dimensional Concrete Frames

Subjected to Corner Column Removal
Yu Jun1 and Song Xuliang2
1
College of Civil and Transportation Engineering, Hohai University, China (yujun@hhu.edu.cn)
2
Guangzhou Metro Design & Research Institute Co. Ltd, China (songxuliang@dtsjy.com)
As progressive collapse is a dynamic procedure in nature, the nonlinear dynamic responses of

three-dimensional (3d) structures are more direct to reflect progressive collapse performance. Among
column removal scenarios, the removal of a corner column must involve the structural performance of
3d structures. Therefore, in this paper, a one-fifth scaled 3dplain concrete frame, consisting of two
stories, two spans and one bay, were built and tested under a corner column removal scenario. The
experimental results are used to demonstrate the collapse modes and to validate numerical models based
on ANSYS/LSDYNA, in which concrete is modeled with solid elements. The validated numerical
models were further employed to investigate the nonlinear dynamic performance of 3d reinforced
concrete (RC) frames subjected to corner column removal, including the time histories of displacements
at critical points, load transfer mechanisms that help to avoid progressive collapse, and load
redistribution in columns after the rebalance of the frame. The results suggest that Virendeel action of
the beams above the removed column was the primary load transfer mechanism, and the substructure
above he removed column induced an overturning moment, resulting in the axial compression in the
two columns adjacent to the removed one significantly increased.
Keywords: Progressive collapse; Concrete frame; Corner column removal
170

1130 - 1150 hours
Feasibility of FRP Composite Material for Building Protection against

Kai Qian1* and Bing Li2
1
Institute of Catastrophe Risk Management, Nanyang Technological University, Singapore
*
(qiankai@ntu.edu.sg)
2
Natural Hazard Research Center, Nanyang Technological University, Singapore (cbli@ntu.edu.sg)
The catastrophic consequences of historic events had indicated that government buildings and
important commercial buildings may need to design in resisting progressive collapse. For newly
designed buildings, they could change the layout, dimensions and reinforcement details of key
members to satisfy the requirements for progressive collapse prevention. However, for existing
structures, retrofitting the members by applying FRP composite sheets or strips is one of an alternative
method as FRP composites materials have the advantages of non-corrosive, non-magnetic, light, strong,
and versatile. In this study, the feasibility of FRP composite material to retrofit structures against
progressive collapse is evaluated by experimental and analytical analysis. Recommendations for design
FRP retrofitting schemes were given based on the observed failure modes and load redistribution
capacities.
Keywords: Progressive collapse; FRP; Tie-force; Membrane action; Reinforced concrete
171

1150 - 1210 hours
Assessment of Progressive Collapse Induced by Blast using Updated Single

Anh Tuan Phama and Kang Hai Tanb

a
(atpham@ntu.edu.sg), b(ckhtan@ntu.edu.sg)
Single column removal scenarios for assessing structural resistance under progressive collapse
have been criticized to be un-realistic when the initial event involves explosions. The assumption of
surviving structures remaining undamaged after the blast event intuitively leads to an unsafe design
methodology. Besides, in a blast event, there may be more than one column damaged in the process. In
an attempt to couple the effects of blast loads on structures when dealing with progressive collapse, this
paper presents the idea of a semi-empirical approach that provides a practical assessment method using
updated column-removal scenarios. Blast effects on remaining structures are taken into account by
employing several scenarios of structural response based on numerical predicted failure modes to
define the limitations of the column-loss assumption. Based on numerical studies validated by test data,
the proposed approach will provide semi-empirical tools such as pressure-impulse diagrams and
analytical equations for RC structural analysis against progressive collapse.
Keywords: Progressive collapse; Contact detonation; Single column removal
172

1210 - 1230 hours
Local Damages of RC Beams According to Variation of Shear

Reinforcement after Blast Loading
Jin-Young Lee1a, Dan-Bi Kim1b, Kyoung-Hwan Min2 and Young-Soo Yoon1c
1
School of Civil, Environmental and Architectural Engineering, College of Engineering, Korea
University, South Korea
a
(jinyounglee@korea.ac.kr), b(tia1107@korea.ac.kr), c(ysyoon@korea.ac.kr)
2
Department of Civil Engineering, Chungcheng University, South Korea (alskh@ok.ac.kr)
In this study, standoff detonation tests and static beam tests on 160×290×2200mm RC beams were
conducted to investigate the effect of local damages caused by blast load on the flexural strength and
ductility index. Seven specimens were manufactured with variables of shear reinforcement ratio and
rebar spacing. New-emulite150, emersion explosive with TNT equivalent of 1.01, was used in blast
test. Local damages of specimens were evaluated by measuring crater, spall diameters and weight loss.
In addition, Flexural strength and ductility index were measured by conducting the static beam tests on
the damaged and undamaged specimens. In these test results, decreasing rebar spacing is the effective
method to increase blast resistance when the shear reinforcement ratio is fixed.
Keywords: Impact; Blast test; Local damage; Residual flexural strength
173

1330 - 1350 hours
Experimental and Theoretical Analysis on the Progressive Collapse

Resistance of One-Way Reinforced Concrete Beam-Slab Substructures
Yulong Zhou1, Peiqi Ren2a, Yi Li2b and Xinzheng Lu2c
1
School of Civil Engineering, Yantai University, China (zhouyulongno.1@163.com)
2
Key Laboratory of Civil Engineering Safety and Durability of Ministry of Education, Tsinghua
University, China
a
(rpq12@mails.tsinghua.edu.cn), b(yili@bjut.edu.cn), c(luxz@tsinghua.edu.cn)
The floor systems consisting of beams and slabs are the primary substructures of frames to resist
progressive collapse. In this work, four 1/3-scaled one-way reinforced concrete (RC) beam-slab
substructures were tested under the middle column removal scenario. The effects of beam height, slab
thickness and seismic design intensity on the progressive collapse resistance were studied. The test
results show that under small deflection, the progressive collapse resistance is mainly provided by the
flexural capacity and the compressive arch action of the substructure; while under large deflection, it is
mainly provided by the catenary tensile force of the reinforcement. The resistance under small
deflection is greatly improved by increasing the beam height; while the resistance under large deflection
is obviously improved by increasing the amount of the reinforcement in the slab. Based on the
experimental analysis of the failure mechanism of the substructures, a theoretical model of the
progressive collapse resistance of RC beam-slab substructures under catenary mechanism was
proposed. Finally, the accuracy of the proposed model was validated by the experimental results of
different researchers.
Keywords: One-way beam-slab substructure; Progressive collapse resistance; Catenary mechanism
174

1430 - 1450 hours
Extended Beam Element Formulation Based on Static Condensation to

Simulate Rotational Discontinuity
Xu Long1*, Kang Hai Tan2a and Chi King Lee2b
1
School of Mechanics, Civil Engineering and Architecture, Northwestern Polytechnical University,
China *(xulong@nwpu.edu.cn)
2
a
(ckhtan@ntu.edu.sg), b(ccklee@ntu.edu.sg)
Rotational discontinuity is commonly encountered in beams, such as pin connection, plastic hinge
and different materials along the length. Due to the fundamental limitation of continuous shape
function, conventional finite element (FE) formulations are not ideal to simulate such rotational
discontinuity. In the last two decades, different approaches have been proposed to extend conventional
finite element formulations by modifying the shape function to incorporate discontinuity properties.
Unlike previous approaches with additional degrees of freedom (DOFs) of unclear physical meanings,
extensive computational cost or fundamental program restructure, static condensation is elegantly
adopted at the level of degree of freedom in this paper to simulate different types of rotational
discontinuity. After validating against several well-established numerical examples, it is found that the
proposed method can be simply incorporated into existing FE-based program with sufficient accuracy
and robustness.
Keywords: Extended beam element; Rotational discontinuity; Plastic hinge; Static condensation
175

1450 - 1510 hours
Bond Characteristics of Concrete-To-SHCC Interfaces and Numerical

Analysis of SHCC-Strengthened RC Slabs under Blast Loading
Lado Riannevo Chandraa, Satadru Das Adhikaryb and Khim Chye, Gary Ongc

a
(ceelrc@nus.edu.sg), b(ceesda@nus.edu.sg), c(ceeongkc@nus.edu.sg)
The retrofit of an existing protective structure against blast loading may require the installation of
new overlay layers on the outer surfaces. If both layers are in direct contact, full knowledge of the bond
characteristics at the interface between the newly installed and the existing substrate layers is necessary.
This paper presents the preliminary results of a study conducted on using strain-hardening cementitious
composites (SHCC) as an overlay material targeted to enhance the resistance of existing reinforced
concrete (RC) flexural panels against blast loading. An experimental study was first performed to
evaluate the bond strength and friction coefficients at the interface of composite specimens cast in the
laboratory. The bond strength between the concrete substrate and new SHCC overlay layers was
assessed through core pull-off tests and modified direct shear tests. To determine the static and dynamic
friction coefficients, a testing procedure similar to that in ASTM-C1028 and ASTM-D1894 was
followed. Experimental results show that concrete substrate surface treatment prior to application of the
SHCC overlays significantly affects the interfacial bond behavior. Finally, to evaluate the feasibility of
SHCC for use as a retrofit material for protective structures, numerical analysis of SHCC-strengthened
RC slabs subjected to blast loading was carried out incorporating the bond and friction parameters
obtained experimentally. Comparative performance assessment study reveals that SHCC-strengthened
slabs perform better vis-à-vis un-retrofitted slabs under the same blast loading conditions.
Keywords: Bond; Concrete; SHCC; Blast loading; Retrofit; Numerical analysis
176

1430 - 1450 hours
Blast Performance of RC Slabs Subjected to Adjacent and Contact Blast

Marek Foglar1*, Martin Kovar1, Radek Hajek1, Josef Fladr1, Jiri Pachman2 and Jiri
Stoller3
1
Czech Technical University in Prague, Faculty of Civil Engineering, Czech Republic
2
University of Pardubice, Faculty of Chemical Technology, Institute of Energetic Materials, Czech
Republic
3
University of Defence, Faculty of Military Technology, Czech Republic
*
(marek.foglar@fsv.cvut.cz)
The author team performed in the past years a set of experiments focused on blast performance of
concrete slabs subjected to adjacent blast loading. The study incorporated 11 slabs in total, dimensions
1.5x0.3x6m, the slabs were made of reinforced concrete. The reinforcement was designed as if the slab
was a bridge deck for the defined span. The slabs differed in concrete compressive strength and in
material and volume of fibers: none, 4.5 or 9kg/m3 poly-propylene fibers, 40 or 80 kg/m3 steel fibers.
The strength of concrete varied from 35 to 90MPa. The steel fibers were made of waste steel,
fy=400MPa.
The 2014 tests focused on blast performance of slabs with 40 or 80 kg/m3 of high strength steel
fibers, fy=2000MPa. Together, slabs made of UHPFRC were tested, one of the specimens had a
glass-fiber matrix inserted in the reinforcement cover. Together with the previous sets of experiments,
the number of tested slabs is 17.
The paper presents the results of the set of experiments and its comparison on spall and breach
curves. The paper formulates conclusions on the blast performance of fiber reinforced concrete.
Keywords: Blast; Concrete; UHPFRC
177

1450 - 1510 hours
Effect of Torsion on 2-D RC Frame under Middle Column Removal in

Namyo Salim Lima, Chi King Leeb and Kang Hai Tanc

a
(nslim1@e.ntu.edu.sg), b(ckhtan@ntu.edu.sg), c(ccklee@ntu.edu.sg)
Many studies in Progressive Collapse had showed that when an RC frame or beam is subjected to
middle column removal scenario, Alternative Load Path (ALP) mechanisms may develop to
redistribute the load to the surrounding members. In 2-D RC frames, these ALPs i.e. Compressive Arch
Action (CAA) and Catenary Action (CAT) are beneficial to the structures as they have higher
resistances than the flexural capacity and may survive until a very large deflection stage under
permissible condition. However, all the tests conducted so far, which show the successful development
of ALPs, are only subjected to in-plane loading (in-plane bending and shear), in the absence of torsion.
In an actual progressive collapse scenario, the exact location of column removal and how the load will
be acting is hard to be controlled. Under certain condition, torsion may be generated on the beam.
Generally, it is believed that torsion will weaken the resistance of structural elements. Hence, the
objective of this research is to study whether the presence of torsion may hinder the development of
ALPs in 2-D RC Frames. In this paper, the test results of two specimens, one subjected to in-plane
loading and another to both torsion and in-plane loading are compared and presented to investigate the
effect of torsion in development of ALPs.
Keywords: Alternative load path; Experiment; Frame; Progressive collapse; Torsion
178

1050 - 1110 hours
Dynamic Behaviour of UHPC Mixed with Nano Material

Yu Su1, 2, Jun Li1, 2, Chengqing Wu1, 2,*, Pengtao Wu1 and Zhongxian Liu1
1
TCU-UA (Tianjin Chengjian University-University of Adelaide) Joint Research Centre on Disaster
Prevention and Mitigation
2
School of Civil, Environmental and Mining Engineering, the University of Adelaide, Australia
*
(j.li@adelaide.edu.au)
Concrete is widely used as a construction material in modern society, however its vulnerability to
brittle tensile failure is becoming a significant drawback when there is a growing need of structural
protection against dynamic loads like seismic or blast in modern environment. To catering such a need,
ultra-high performance concrete (UHPC) which is known for its high compressive and tensile strength
as well as ductility and durability is under fast development. In a recent study, an innovative UHPC
material is developed. The addition of nano size material in the composite matrix differentiates it from
conventional ultra-high performance concrete. Split-Hopkinson Pressure Bar (SHPB) tests are
conducted on this new concrete material to obtain its dynamic compression properties. In this paper, the
effect of different nano materials addition and volume fraction on the behaviour of UHPC is
summarised. Dynamic increase factors for this material are presented based on comparison between
static tests and SHPC tests.
Keywords: UHPC; SHPB
179

1110 - 1130 hours
Experimental Frequency Domain Assessment of Direct Shear in NSC and

UHPC
Earl Maher1*, Theodore Krauthammer1, Michael Stone1, Robin French2, Jaeyoon Kim3
1
Center for Infrastructure Protection and Physical Security, University of Florida, USA
*
(earlmaher@ufl.edu)
2
1 Engineer Support Unit, Canadian Army, Canada
3
Civil and Environmental Engineering, University of West Virginia, USA
Direct shear is recognized as a possible response mechanism in structural concrete systems

subjected to severe dynamic loads, and it can lead to catastrophic failure. Behavioral models for direct
shear in normal strength concrete (NSC) were introduced in the 1970s and their adaptation for the
analysis of structural response under blast and ground shock effects was presented in the 1980s. The
introduction of ultra-high performance concrete (UHPC) for protected facilities has created the need to
reevaluate direct shear in both NSC and UHPC, and to characterize this response mechanism more
accurately. Therefore, direct shear impact tests were conducted on NSC and UHPC specimens with
three varying reinforcement ratios and the results were analyzed in both the time and frequency
domains. This paper is focused on the assessment of direct shear in the frequency domain to identify the
relationships between parameters of the direct shear resistance functions for NSC and UHPC, and the
power spectral density of the captured data from impact tests.
Keywords: NC; NSC; UHPC; Direct Shear; Impact; Frequency Domain; Impulsive; Protective Design
180

1130 - 1150 hours
Response to RC Slabs Subjected to Combined Blast and Fragment Loading

Rickard Forsén
FOI, Swedish Defence Research Agency, Sweden

(rickard.forsen@foi.se)
Damages to building structures caused by conventional weapons such as aircraft bombs and
artillery shells, but also improvised explosive devices such as car bombs, depend both on the blast
pressure and on the fragments that are generated. In many cases it is important to evaluate the
vulnerability or protection performance of both ordinary unprotected structures housing vital societal
functions and shelter structures that may be subjected to conventional weapons effects. A large number
of tests have been performed to determine the damage to RC (reinforced concrete) slabs subjected to
fragment loading with different velocities, fragment masses and mass densities over the slab surface.
After the fragment loading the slabs were statically loaded to determine their residual load bearing
capacity. The damaged slabs will lose part of its load bearing capacity making them act as slabs with
reduced thickness. Relations between fragment load characteristics and effective thickness reduction
were evaluated. Based on the results a fast running computational tool using single degree of freedom
analysis has been developed. The tool considers the bending mode of the slabs and makes it possible to
determine the maximum deflection due to combined blast and fragment loading. This paper summarizes
the experimental work to determine effective slab thickness reduction and the algorithms used in the
fast running tool. It also presents some comparisons between calculations and results from experiments
with combined blast and fragment loading.
Keywords: Blast; Fragments; Reinforced Concrete
181

1150 - 1210 hours
End Restraint Effect on the Dynamic Response and Failure Behavior

of Steel Beams Subjected to Pulse Loading
Feng Xi*, Yueqiang Li, Deyang Kong, Yu Zhang
*
School of Civil Engineering, Shandong Jianzhu University, China
(xifeng@sdjzu.edu.cn)
This paper deals with the generic modeling of a steel member subjected to pulse loading where the
steel element is restrained at its ends by translational and rotational springs. The rate-dependent effect
as well as the failure effect of material are considered in proposed model. The effects of the end restraint
on the dynamic response and failure of steel beams are fully investigated. Analysis results indicate that
in the cases of only axial restraint or rotational restraint, the displacement reduce considerably as the
restraint stiffness increase, while in the both cases of the end with complete axial or rotational restraint,
the corresponding impact on the response is not serious with increase of rotational or axial restraint
stiffness, respectively. A 3-D diagram on the relationship between the failure impulse with both axial
and rotational restraint stiffness is also presented, which shows that the explosion resistance of beam
basically increase with axial restraint stiffness increase or rotational restraint stiffness decrease, and the
minimum failure impulse emerges obvious edge effect when restraint stiffness is low.
Keywords: Restrained beam; Larger deformation; Failure; Dynamic response; Computing model
182

1210 - 1230 hours
An Assessment Protocol for Reinforced Concrete Columns Subject to

Internal Building Detonations
Wijesundara M.G. Lakshitha* and Simon K. Clubley
Faculty of Engineering and the Environment, University of Southampton, United Kingdom

*
(glk2g10@soton.ac.uk)
This paper details the development of an engineering assessment procedure for reinforced concrete
column failure due to time-variant coupled uplift forces and lateral blast pressures characteristic of
internal building detonations. This is primarily based on ratified high-resolution computational fluid
dynamics simulations and advanced transient structural dynamic modelling techniques. A set of
comprehensive parametric studies are conducted and results are analyzed using a non-linear curve
fitting technique to develop assessment charts quantifying the extent of overall damage and residual
axial column capacity following internal blast actions.
Keywords: Internal explosions; Impulsive; Protective design; Numerical simulations
183
Structures under Fire

STRUCTURES UNDER FIRE ABSTRACTS

1330 – 1350 hours
Application of Cellular Automaton in Fire Engineering
Qing Ying and Weifeng Yuan*
China
*
(yuanweifeng@swust.edu.cn)
Cellular automaton (CA) is a discrete method which is used to simulate complex system. In this
study, CA is extended to the field of fire engineering. First, a numerical algorithm is developed to
calculation the heat conduction in composites. A composite concrete column is analyzed as an example
to demonstrate the reasonability of this algorithm. Second, a comprehensive CA model which takes
human behaviour into account is proposed in the simulation of emergency evacuation in fire. Several
numerical examples show that this CA model is versatile and it has great potential to be used in fire
safety. At last, a CA-based method that can be used to model the deformation of 2-dimensional elastics
is introduced. This method is a meshless algorithm and it has very good parallel calculability.
Keywords: Cellular automaton; Heat conduction; Evacuation; Meshless
186

1350 – 1410 hours
Using Performance-based Fire Engineering Design in a Protected Facility
Yong Kiang Tan a , Chee Leong Ho b and Yew Soon Ng c

a
(rick_tan@dsta.gov.sg), b(ho_chee_leong@dsta.gov.sg), c(ng_yew_soon@dsta.gov.sg)
When designing large and complex buildings, fire safety engineers in Singapore are moving away
from prescriptive-code compliance to performance-based fire safety design. This move is motivated by
economics, as well as technical and architectural considerations.
Performance-based fire safety design opens up new possibilities for optimising protective facility
design. Within large buildings, there are protected spaces designed to be used by a limited group of
military personnel who have to be protected from potential hostile environments. In such protected
spaces, fire safety provisions can be specially integrated and added to the military protection concept.
This paper outlines the fire safety design methodology for such large buildings.
The paper provides an illustration of a performance-based approach using a fictitious protected
facility. In this instance, the openings required for the smoke extraction system were optimised so that
these ventilation openings met both fire safety and military protection requirements. The illustration
also demonstrates the application of Fire Dynamics Simulator, which is a computational fluid dynamics
(CFD) simulation tool, to determine the required smoke extraction rate. The performance-based
approach illustrated also demonstrates acceptance criteria for fire and smoke control and for evacuation
can be derived in jointly by designers with the building’s occupants and the fire authorities. Finally, the
way the building performs under different fire scenarios is assessed against the fire authority’s fire
safety objectives, as well as the acceptance criteria established earlier.
Keywords: Performance based fire safety design; Protective facility
187

1410 – 1430 hours
Analytical Evaluation of Bi-Directionally Prestressed Concrete Panel

Behavior under Blast-Fire Combined Loading using Parametric Study
Seung-Jai Choia, Dong-WooShinb, Jae-Ho Shinc and Jang-Ho Jay Kimd
School of Civil and Environmental Engineering, Yonsei University, South Korea

a
(seungjaechoi@yonsei.ac.kr), b(patshin@naver.com), c(naeshkill@hanmail.net),
d
(jjhkim@yonsei.ac.kr)
In recent years, numerous explosion and collision related tragedies due to military attack, terrorist
bombing, and vehicle accident have occurred all over the world. However, researches on Prestressed
Concrete infrastructures such as Prestressed Concrete Containment Vessels (PCCVs), and LNG storage
tanks under extreme loading such as blast induced fire loading scenario are not being studied in depth.
Therefore, in this study, the bi-directional concrete panels such as walls of PCCVs and LNG storage
tanks were experimentally and analytically evaluated under blast-fire combined loadings. Blast
resistance tests were using ANFO 35 lbs. Also, blast induced fire loading resistance was tested using
RABT fire loading scenario. Then, blast and fire behaviors of the PSRC panels were evaluated using
finite element analysis program of LS-DYNA and MIDAS FEA, respectively. The simulation results
from the PSRC specimen applied with blast-fire combined loadings were compared with those of
unblasted PSRC specimens. For the simulation of blast induced fire loading, RABT fire load was
applied to the blast damaged panel model based on LS-DYNA simulation. The reliability of the analysis
results were evaluated by performing parametric study for variables of compressive strength,
reinforcement ratio, prestressing force, and cover thickness.
Keywords: Blast-fire combined loading; Parametric study; Concrete Panel; Analytical evaluation
188

1430 – 1450 hours
Effect of Braces on the Progressive Collapse Mechanism of Steel Structures

Exposed to Fire
Jian Jiang1, Guo-Qiang Li2 and Asif Usmani3
1
College of Civil Engineering, Tongji University, China (jiangjian_0131@163.com),
2
State Key Laboratory for Disaster Reduction in Civil Engineering, Tongji University, China
(gqli@tongji.edu.cn),
3
School of Engineering, the University of Edinburgh, United Kingdom (asif.usmani@ed.ac.uk)
This paper investigates the effect of various bracing systems on the fire-induced progressive
collapse resistance of steel-framed structures using OpenSees. Two types of bracing systems (vertical
and hat bracing) are considered. Four collapse mechanisms of steel frames in fire are found through
parametric studies. General collapse is characterized by the collapse of the heated bay followed by
lateral drift of adjacent cool bays. Global collapse of a whole frame is due to the buckling of ground
floor columns. Another two lateral collapse modes (local and global) are caused by catenary action
developed in the heated beams under large deflections. Vertical bracing systems have positive effects on
increasing the lateral restraint of the frame against local or global drift, while when arranged at edge
bays of frames they negatively contribute to the spreading of a local damage to global collapse in the
form of sequential buckling of adjacent columns through load-transfer mechanisms. Instead, using hat
bracing can effectively optimize the load-transfer path through a more uniform redistribution of loads in
columns and enhance the resistance of structures against progressive collapse. It is found that the
application of vertical bracing systems alone on the steel frames to resist progressive collapse is proved
to be unsafe and a combined vertical and hat bracing system is recommended in practical design.
Keywords: Bracing system; Progressive collapse; Steel frame; Fire; OpenSees
189

1450 – 1510 hours
Design of Wireless System for the Detection of Accidental and Terrorist

Explosions and Fires in Manufacturing Sites
Edgar Mataradze1, Mikheil Chikhradze1, Theodore Krauthammer2 and Shalva
Marjanishvili3
1
G.Tsulukidze Mining Institute, Georgia (mataradze@mining.org.ge)
2
Center for Infrastructure Protection and Physical Security (CIPPS), University of Florida, USA
(tedk@psu.edu)
3
Hinman Consulting Engineers, Inc., USA (shalva@hce.com)
The development of contemporary detection devices is oriented to the creation of integrated

systems ensuring monitoring of all possible threats as well as respective measures for threat prevention.
A comprehensive security system enables identification of emergency and pre-emergency conditions
and have WiFi modes of operation. The system shall be quick-acting and reliable, and not impede the
normal functioning of the manufacturing sites and/or underground facility. In order to achieve relibility,
the system should not depend on the external power. Detection system have to meet the requirements of
the Directive 94/9/EC on equipment and protective systems intended for use in potentially explosive
atmospheres (ATEX).
This paper describes results of experimental investigations to identify and process signals from
various sources. The described works are necessary for designing the integrated module of
identification which is based on the constant monitoring on potential threat that may take place in
manufacturing sites and tunnels. The potential threats discussed in the paper are:
 the blast, which is based on the monitoring of overpressure and methane concentration control
in the media;
 the fire, which identification is based on the monitoring of flame and smoke parameters.
Besides the reliable identification and generation of the emergency signal, proposed
multifunctional device will ensure also the activation of the protective systems to reduce the damage
and fatal consequences.
Results of tests of system show that potentially it can be used for identification of danger in 2.4 ms
after excitation of a sensor of the detector and activation of protection device in 11 ms after the time of
explosion.
Keywords: Explosion; Fire; Detection; Signal transmitter; Signal Receiver; Wireless system
190

1510 – 1530 hours
Performance of Tall Buildings in Post Earthquake Fire
T.Ch. Madhavi
Department of Civil Engineering, SRM University, India
(tcmadhvi@yahoo.com)
Post-Earthquake fire is a multi-hazard scenario and causes severe damage to structures and also
human life. Generally, tall structures are designed to resist seismic forces and designed to fight fire
separately. But, in the event of both occurring together, the resistance to failure is considerably reduced
as structures are seldom designed for such cases. This important factor is not considered explicitly in the
structural design and codes. In post-earthquake, not only the building, but also its fire protection system
will be damaged and hence the structure fails easily when it is under fire. In some cases, the losses due
to fire are much more than those caused by the earthquake itself and may even result in a catastrophe.
Past records show that the losses due to the fire caused after the occurrence of an earthquake were
significant and at times even catastrophic. Post-earthquake fire has showed the potential to cause
significant damages to tall buildings and hence the building should be safe to resist the fire at least for
minimum time duration so as to allow for safe evacuation. It is essential to develop measures to mitigate
the damages caused by such highly destructive fires. The paper discusses in detail the various factors
associated with fire safety in tall buildings, performance of structures, requirements, evaluation of
performance and evacuation to safety under scenario of fire after earthquake.
Keywords: Post-earthquake Fire; Tall Buildings; Performance evaluation; Fire damage
191

POSTER ABSTRACTS
Progressive Collapse Analysis of High-rise Structures Due to Fire:

A case study on Shanghai Tower
Jian Jiang1, Guo-Qiang Li2, Asif Usmani3
1
College of Civil Engineering, Tongji University, China
( jiangjian_0131@163.com)
2
State Key Laboratory for Disaster Reduction in Civil Engineering, Tongji University, China
(gqli@tongji.edu.cn)
3
School of Engineering, The University of Edinburgh, United Kingdom
(asif.usmani@ed.ac.uk)
Shanghai Tower is a high-rise building composed of mega frame-core-outrigger lateral resisting

systems. Its structural safety due to fire has been paid great attentions. This paper presents
investigations on the performance of Shanghai Tower against fire-induced progressive collapse. The
fire resistance of key components including concrete core, mega columns, composite floor, outrigger
trusses, belt trusses are examined first based on both the standard and real fire scenarios. Their real fire
resistance periods are proved to be far beyond the design fire resistance. The components with weak fire
resistance such as peripheral steel columns and web members of belt trusses are then removed to study
the resistance of the residual structure against progressive collapse. The results show that Shanghai
Tower has good resistance against fire-induced progressive collapse. The concrete components have
smaller residual displacements compared to the steel components. It is recommended that effective fire
protection should be guaranteed for the outrigger trusses to maintain the connection between the core
and mega columns.
Keywords: Progressive collapse; Case study; Fire; Shanghai Tower
194
POSTER ABSTRACTS
Optimal Design of Close-Celled Metallic Foam Protection for Low Velocity

Impact
Bin-Chao Li2, Gui-Ping Zhao1a and Gai-Hui Zhang1b
1
State Key Laboratory for Mechanical Structure Strength and Vibration, School of Aerospace, Xi’an
Jiaotong University, China a(zhaogp@mail.xjtu.edu.cn), b(zhanggh@mail.xjtu.edu.cn),
2
National Key LAB of Science and Technology on LRE, Xi’an Aerospace Propulsion Institute, China
(libinchao@stu.xjtu.edu.cn)
A single-degree-of-freedom colliding model for large-mass structures protected by high porosity

close-celled aluminum foam cushions subjected to low velocity impact is proposed. Upon validating the
model with drop hammer test results and finite element simulations, the maximum acceleration
experienced by the protected structure is minimized under different impact conditions. The influence of
the impact mass ratio, foam porosity, and geometrical dimensions of foam cushion upon the
acceleration is systematically carried out, and it is demonstrated that increasing the foam thickness,
mass ratio or porosity can reduce the acceleration. Finally, the weight and geometrical dimensions of
the foam cushion are optimized under different constraint conditions.
Keywords: Low velocity impact; Theoretical model; Aluminum foam; Optimization
195
POSTER ABSTRACTS
Discrete Optimization of Steel Beams Subjected to Blast Load

Primoz Jelusica and Bojan Zlenderb
Faculty of Civil Engineering, University of Maribor, Slovenia

a
(primoz.jelusic@um.si), b(bojan.zlender@um.si)
This study presents the discrete optimization of steel elements with two different approaches. First,
the discrete optimization is performed with program MATLAB where Exhaustive Enumeration (EE)
method was employed. Second, the optimization was preformed with program SAP2000 using auto
selection list. The approaches are compared in terms of process time and ease-of-use. The computer
code for discrete optimization of simply supported laterally restrained steel beam according to
Eurocode 3 is developed based on the steel sections present in the market according to the European
beams tables. Laterally restrained beams are checked for resistance of cross-section to bending,
resistance of cross-section to shear, resistance to shear buckling, resistance to flange-induced buckling,
resistance of the web to transverse forces and deflection. In the beam design, a practical solution
involves identifying an “of the shelf” beam with minimum weight to support the given load. Unless the
element is to be subjected to repeated blast loading, the design should be based on the ultimate limit
state. However, presented approaches could be applied in the design of blast load-resisting structural
elements.
Keywords: Discrete optimization; Exhaustive enumeration; Numerical modeling; Design of steel beams
196
POSTER ABSTRACTS
Numerical Modeling of Cased Charge in a Pipe Using a Coupled

CFD-CSD Methodology
Joseph D. Baum1a, Orlando A. Soto1b, Fumiya Togashi1c and Rainald Löhner2
1
Applied Simulations, Inc, USA
a
(joseph.d.baum@appliedsimulations.com), b(orlando.a.soto@appliedsimulations.com),
c
(fumiya.togashi@appliedsimulations.com
2
George Mason University, USA (rlohner@gmu.edu)
This paper describes the application of a state-of-the-art coupled computational fluid dynamics
(CFD) and computational structural dynamics (CSD) methodology to the simulation of an
explosive-filled steel cylinder placed inside a long pipe. The physical mechanisms controlling the
explosive detonation and cylinder break-up include: HE detonation initiation, detonation wave
propagation through the HE, cylinder expansion under the high pressure, cylinder cracking, detonation
products escape through the opening cracks, accelerating cylinder debris and air blast impact on the
steel pipe, pipe response and air blast, cylinder and pipe debris propagation down the pipe. Accurate
modeling of these mutually interacting processes requires coupling CFD-CSD codes.
Three simulations were conducted under this study: a bare charge detonation and two identical
HE-in-a-cylinder charges. The first of these two modeled the cylinder debris to bounce elastically from
the pipe; the second modeled the steel pipe response to the blast and debris loading. All models used
identical explosive charges. The initial fluid energy loss required to break the cylinder and accelerate
the cylinder debris is identical for both cylindrical charge simulations. At later times, the high-speed
fragments bouncing elastically from the non-responding pipe accelerate the flow and reinstate a
significant portion of the energy lost. For the responding pipe, most of the impacting fragment energy is
lost on the shock wave within the steel pipe and the plastic work performed. Hence, the slow-velocity
reflected fragments reinstate only a small portion of the fluid energy lost. The reflected cylinder debris
from the pipe periphery focus toward the centerline, and recompress (or squeeze) the high-temperature,
high-pressure core, to form axial jets, waves with large mass and energy (both internal and kinetic)
fluxes.
Keywords: Impact; Fluid-structure Interaction; Structural Response to Blast and Fragments
197
POSTER ABSTRACTS
Plastic Hinge Formation of Beam-Column Connections under Blast Loads

Kwang Mo Lim1, Yong Soo Yoon2 and Joo Ha Lee1b
1
Department of Civil Engineering, The University of Suwon, Republic of Korea
a
(kwangmolim@suwon.ac.kr), b(leejooha@suwon.ac.kr)
2
School of Civil Environmental & Architectural Engineering, Korea University, Republic of Korea
(ysyoon@korea.ac.kr)
Given the prospective threat of unexpected explosion accidents or terrorist attacks causing severe
damage to the private facilities as well as military facilities, it is very important to investigate and
evaluate the blast resistant behavior of structural members. As a particularly critical structural element
under blast loading, beam-column connections are the specific subject for this study. To understand the
dynamic behavior of reinforced concrete beam-column connections, plastic hinge formation has been
studied. Blast loads can induce severe damage with large displacements in regions subjected to large
moments. These regions are commonly referred to as plastic hinges. However, predicting the exact
behavior of plastic hinges is difficult and so previous researches showed varied results. Moreover, most
researches have focused on plastic hinge of flexural members under static or dynamic loads, rather than
under blast or impact loads. In addition, for reinforced concrete beam-column connections designed by
current design recommendations, plastic hinges usually form in the beam region adjacent to the column
face, which can cause structural problems in terms of stiffness and strength in the joint. It can be
beneficial to relocate plastic hinges at a certain distance away from the connection. Therefore, this
analytical study investigated various reinforcement details to relocate beam plastic hinges and other
design parameters affecting the location of the plastic hinge.
Keywords: Blast loads; Concrete structures; Beam-column connection; Numerical analysis
198
POSTER ABSTRACTS
Experimental Study of Square CFDST Columns with Steel-Fiber

Reinforced Concrete under Blast Loading
Fangrui Zhang1, Chengqing Wu1*, 2 and Xiao-Ling Zhao
1
School of Civil, Environmental and Mining Engineering, The University of Adelaide, Australia
*
(cheng.wu@adelaide.edu.au)
2
TCU-UA (Tianjin Chengjian University -- University of Adelaide) Joint Research Centre on Disaster
Prevention and Mitigation
In recent years, a large number of researches have been carried out to investigate the behaviours of
concrete filled double skin steel tube (CFDST) members due to its increasing popularity in the
construction industry. This paper presents an experimental study on the ultra-high performance concrete
filled double-skin tubes subjected to blast loading. The cross-sectional geometries of the tested
specimens were square for both inner and outer steel tubes. Different explosive charge weights and
different axial load levels were applied to investigate their influences on the structural behaviours,
respectively. After the test, there were no visible buckling nor ruptures found on the steel tubes and only
minor cracks of no more than 1mm width were observed in the core concrete when the outer steel tube
was removed. Based on the findings, it is evident that the proposed CFDST column has a great blast
resistance and this feature has the potential to be used in high-value structures which maybe the targets
of terrorist attacks, such as embassies, government buildings and critical infrastructures.
Keywords: Steel-fiber reinforced concrete; CFDST members; Blast Loading
199
POSTER ABSTRACTS
Dynamic Response of Geometrically Asymmetric Metal Sandwich Plates

Subjected to Blast Loading
Mingshi Wang1, Tao Wang2, Qinghua Qin3*, Zhengjin Wang4, T. J. Wang5*
1
(mingshiwang@stu.xjtu.edu.cn)
2
Mechanics, Xi’an Jiaotong University, China; Department of Engineering Physics, Xi’an Research
Institute of High-Tech., China
(56448276@qq.com)
3*
State Key Laboratory for Strength and Vibration of Mechanical Structures, Department of
Engineering Mechanics, Xi’an Jiaotong University, China
4
(635501834@qq.com)
5
*
(wangtj@mail.xjtu.edu.cn)
The dynamic response of geometrically asymmetric metal sandwich structures subjected to blast
loading is investigated theoretically and experimentally. The fully clamped square metal sandwich
plates with geometrically asymmetric cross-sections are considered. The yield criteria for geometrically
asymmetric compressed sandwich cross-sections are presented. The dynamic response of the
asymmetric sandwich plates is decoupled into core compression phase and plastic bending/stretching
phase in theoretical analysis. Using the circumscribing and inscribing yield loci, the so-called ‘bounds’
of the maximum central deflection of the back face-sheet of the geometrically asymmetric sandwich
plates are derived. Honeycomb core sandwich specimens with different ratios of front face-sheet
thickness to back face-sheet thickness subjected to different levels of blast loading were tested. It is
found that the theoretical predictions are in good agreement with the experimental results.
Keywords: Blast loading; Metal sandwich plate; Geometric asymmetry; Dynamic plasticity; Large
deflection.
200
POSTER ABSTRACTS
Performance Fiber Reinforced Cementitious Composites under Mode I

Fracture
Seungwon Kim, Hoki Ban, Minkwan Ju, Hyunmyung Cho, and Cheolwoo Park*
Department of Civil Engineering, Kangwon National University, South Korea

*
(tigerpark@kangwon.ac.kr)
High Performance Fiber Reinforced Cementitious Composite (HPFRCC), with its high strength,
ductility and toughness, can be successfully applied to the blast resistant structures. The fiber networks
in the matrix, with fiber volume fractions of 4% to 20%, provide high resistance to impact or blast
loading. The fracture mode of HPFRCC subjected to the impact or blasting, may reveal in general, a
mixed mode, as well as complicated crack growth. However, the HPFRCC with high volume fraction of
steel fiber, exhibits somewhat homogeneous responses, even though there is a certain distortion at the
right front of crack initiation. Better in-depth understanding on the fracture behavior of the high volume
fraction HPFRCC, may produce better predictions on the structural performance. To this end, integrated
experimental and numerical efforts were made to characterize the fracture behavior of HPFRCC. In
experiments, beam flexural fracture tests were performed to demonstrate pure mode I (opening). The
test configuration for mode I test had a notch depth of 25 mm in the middle of specimen. During the
tests, vertical displacements were measured with LVDT. Crack mouth opening displacements (CMOD)
were measured with clip-on-gages. These were used to compute the fracture energy. The specimens
used in this study consisted of a cementitious matrix of Portland cement, silica fume and fine aggregate
and 8% volume fraction of steel fibers. Along with the laboratory tests, numerical simulations were
performed to obtain fracture properties of HPFRC. Numerical simulations were compared with the
laboratory test results. In the modeling, the cohesive zone model was incorporated to predict the
progressive fracture at the notch tip.
Keywords: Fracture resistance; High Performance Fiber reinforced Cementitious composites

(HPFRCC); Mode I Fracture
201
POSTER ABSTRACTS
Impact Load Response of Reinforced Concrete Beams with Hsswm-Hpm

Retrofitting
Weizhang Liao1*, Miao Li1, Wei Zhang2, Zhimin Tian3 and Shuangjun Shen1
1
School Civil and Transportation Engineering, Beijing University of Civil Engineering and
Architecture, China
2
Franshion Properties (China) Limited, China
3
The Fourth Engineering Scientific Research Institute of the Headquarters of the General Staff,
China
*
(liaoweizhang@bucea.edu.cn)
High Strength Steel Wire Mesh and High Performance Mortar (HSSWM-HPM) is widely used as
a viable method for structural strengthening. Recent researches suggest that HSSWM-HPM can be
applied for enhancing earthquake-resistance of structures. The research reported herein evaluates the
mechanical performance of reinforced concrete (RC) beams with the HSSWM-HPM in drop-weight
impact tests using both laboratory experiment method and numerical simulation method. In the
laboratory experiments, there are eight full-size beams, including five RC beams and three unreinforced
beams for comparison. Throughout the drop-weight impact test, the following parameters were
recorded: including the impact loads, strains, accelerations, velocities and deflections of the beams, and
residual damage was documented. The performances of eight beams are compared with each other.
Another comparative study on the failure pattern and the impact resistance of beams with and without
reinforcement were carried out using the finite element method in the explicit analysis software,
LS-DYNA. A general prediction of time history and maximum mid-span deflections of the beams can
be acquired from LS-DYNA.
The research results obtained from both laboratory experiments and finite element analyses
indicate that the impact resistance properties and ductility of the reinforced RC beams are significantly
improved.
Keywords: Impact loads; Drop-weight impact test; Explicit analysis software
202
POSTER ABSTRACTS
Experimental Study of Failure Damage on Prestressed Concrete Panels

under Impact-Fire Combined Loading
Na-HyunYi1, Ji-HunChoi2a, Sang-Won Lee2b, Jang-Ho Jay Kim2c
1
(NHYi@ntu.edu.sg)
2
School of Civil and Environmental Engineering, Yonsei University, Korea
a
(cjh228@yonsei.ac.kr), b(lsw204@naver.com), c(jjhkim@yonsei.ac.kr)
Although impact, terror, fire, and explosion incidents have increased globally, existing design
procedures for many civil infrastructures still do not consider a protective design for extreme loadings
from blast, impact, and fire. Such considerations are especially lacking in the design of major
infrastructure such as bridges, tunnels, PCCVs (Prestressed Concrete Containment Vessels), and LNG
storage tanks. Many of these infrastructures use Prestressed Concrete (PSC) systems. There is a lack of
research into PSC systems behavior under such extreme loads to provide knowledge to support design.
Beyond the immediate effects of blasts and impact from bombings, collisions of vehicles and
vessels from terror attacks, there are also possible secondary effects from fire. Thus in an attack, several
effects may affect structures ,either together, or one after another in rapid succession, rather than in
isolation. Most past research has not focused on the combined effects of the extreme loads. The research
has instead focused on isolated extreme loadings and their effects on the structures.
Therefore, in this study, experiments were conducted on bi-directional prestressed concrete panels
to subjected first to loads from impact, and then to fire. Each PSC panel was 1400×1000×300 mm in
dimension, with 430kN prestressing force applied using unbonded prestress thread bars. Ordinary RC
structures were also studied to provide a comparison against the PSC panels. For each panel impact
resistance was first evaluated by using a 14kN impactor, dropped from a height of 3.5m.The impacted
panels were then subjected to fire, following the RABT fire loading curve. The RABT fire curve is a
thermal loading curve where temperature is raised to 1200Ԩ within 5 minutes. Finally the residual
structural strength of PSC and ordinary RC panels was assessed and compared. The experimental
results provide data for research in protective design and for calibrating numerical simulations.
Keywords: Prestressed concrete; Impact-fire combined loading; RABT fire scenario; PCCV
203
POSTER ABSTRACTS
Analytical Evaluation of Bi-directionally Prestressed Concrete Panel

Behavior under Impact-Fire combined Loading using Parametric Study
Sang-Won Leea, Cheol-Min Chob, Tae-Kyun Kimc and Jang-Ho Jay Kimd
School of Civil and Environmental Engineering, Yonsei University, Korea

a
(lsw204@yonsei.ac.kr), b(mngt0313@nate.com), c(ssida24@naver.com),
d
(jjhkim@yonsei.ac.kr)
In recent years, numerous explosion and collision related tragedies due to military attack, terrorist
bombing, and vehicle accident have occurred all over the world. However, researches on Pre-stressed
Concrete infrastructures such as Pre-stressed Concrete Containment Vessels (PCCVs), and LNG
storage tanks under extreme loading such as impact induced fire loading scenario are not being studied
in depth. Therefore, in this study, the bi-directional concrete panels such as walls of PCCVs and LNG
storage tanks were experimentally and analytically evaluated under impact-fire combined loadings.
Impact resistance tests were performed with 14kN impactor. Also, impact induced fire loading
resistance was tested using RABT fire loading scenario. Then, impact and fire behaviors of the PSC
panels were evaluated using finite element analysis program of LS-DYNA and MIDAS FEA,
respectively. The simulation results from the PSC and RC specimens subjected to impact-fire combined
loadings were compared with those of unimpacted PSC and RC specimens. For the simulation of impact
induced fire loading, RABT fire load was applied to the impact damaged panel model based on
LS-DYNA simulation. The reliability of the analysis results were evaluated by performing parametric
study for variables of compressive strength, reinforcement ratio, pre-stressing force, and cover
thickness.
Keywords: Pre-stressed concrete; Impact-fire combined loading; Parametric study; RABT fire scenario
204
POSTER ABSTRACTS
A Simplified Analysis Method on Reinforced Concrete Columns

Subjected to Uniaxial Bending in Fire
Truong-Thang Nguyen1 and Kang Hai Tan2
1
Department of Concrete Structures, National University of Civil Engineering, Vietnam
(thangcee@gmail.com)
2
(CKHTAN@ntu.edu.sg)
This paper introduces a simplified method for cross-sectional analysis of reinforced concrete (RC)
columns under the simultaneous impacts of axial force, bending moment about one principal axis of the
cross-section (so-called uniaxial bending) and standard fire conditions. Based on a rational approach for
fire-resistant analysis of RC columns developed from the specifications of EC2 Parts 1.1 and 1.2 and
validated by a fair number of fire tests, this simplified method proposes the novel terms of equivalent
concrete compressive stress block (with equivalent factors of  and ) and an equivalent temperature
s of longitudinal reinforcing bars for the cross-sectional analysis of the column at elevated
temperatures. Practical equations and design charts are generated. By conducting a series of parametric
studies, the effects of standard fire exposures, longitudinal bars, concrete cover, concrete compressive
strength, aggregate types and cross-section sizes are investigated. Then, the temperature-dependent
load resistance of RC columns can be expressed in the form of regressing M-N interaction diagrams,
based on which a number of design recommendation for RC columns subjected to uniaxial bending
under standard fire conditions can be addressed. This method is capable for using the Eurocodes’
specifications to obtain better understanding in the analysis of RC structures at elevated temperatures so
that the performance-based approach for structural fire engineering to the Eurocodes can be further
developed.
Keywords: Reinforced concrete; Structure; Column; Fire; Eurocde
205
POSTER ABSTRACTS
Numerical Simulation on Propagation of Shock Wave at Subway Entrances

under Internal Blast Loading
Hejia Wang1*, Weizhang Liao1 and Zhimin Tian2
1
School Civil and Transportation Engineering, Beijing University of Civil Engineering and
Architecture, China
2
The Fourth Engineering Scientific Research Institute of the Headquarters of the General Staff,
China
*
(wanghj0404@foxmail.com)
Subways have become one of the targets of terrorist attacks using explosives. Explosion in
subway structure may cause the damage of them and lead to further loss of lives and properties. The
study motivated by the fact that the security checks at the entrances of subways structure are not able to
avoid terrorist attacks completely. The paper aims to analyze the propagation of blastshock wave at
subway entrances under the internal blast loading, and investigates the influences of various factors on
propagation law, including layout of the entrance, the explosive detonation location and the weight of
explosive. A three-dimensional finite element model including explosive detonation, air and subway
entrances structure are established to address the issue on the shock wave waveform, attenuation and
walking pattern using the explicit finite element software LS-DYNA. It is found from the numerical
simulation study that the entrances layout would affect propagation of shock wave very well. The
results provide the theoretical basis for further experimental study.
Keywords: Subway entrances; Propagation; Numerical simulation; Shock wave
206
POSTER ABSTRACTS
Computational Modelling of Pre-Stressed Confined Ceramic Armor

Structures
Ahmad Serjouei*, Xianfeng Zhang and Idapalapati Sridhar
1
School of Mechanical & Aerospace Engineering, Nanyang Technological University, Singapore
*
(ahma0015@e.ntu.edu.sg)
In this work, a numerical model consisting of six steps for simulation of pre-stressed ceramic
targets under high velocity impact, using explicit software, AUTODYN®, is developed. Impact
simulations of confined ceramic targets with and without pre-stress are carried out. Effect of pre-stress
in radial, axial and hydrostatic status with different intensities on ballistic performance of the ceramic
armor is explored. It is shown that hydrostatic pre-stress is more effective than radial pre-stress (and
also axial pre-stress) in improving the ballistic behavior of the thick ceramic armor. Finally, it is shown
that, for thin ceramic targets, cone cracks are formed and with increasing of pre-stress value, the cone
angle decreases.
Keywords: Pre-stress; Ballistic performance; Thick and thin ceramic armor
207
POSTER ABSTRACTS
Modelling Tm5-855-1 Peak Pressure Plot Using LS-DYNA

Eng-Choon Leong1a, Jia Han Chew1b and Sarma Anand2
1
a
(cecleong@ntu.edu.sg), b(jchew008@e.ntu.edu.sg)
2
Defence Science & Technology Agency, Singapore (sanand@dsta.gov.sg)
Calibration of numerical model is an intrinsic part of a numerical study to model a field

phenomenon. In ground shock studies, this will mean that a number of field tests are required to
calibrate the numerical model. The peak pressure plot for various soil types presented in TM5-855-1[1]
was developed from a database of field tests. It has served as a useful guide for estimates of free-field
peak pressure. Coupling this free-field peak pressure with a reflection factor, a peak pressure loading on
an underground structure can be estimated. The objective of this paper is to present a calibration of a
material model in LS-DYNA for modelling of the peak free-field pressure plot in TM5-855-1[1]. The
calibration results showed that it is possible to capture the correct peak free-field pressure responses for
various soil types with material model MAT 5 in LS-DYNA. The calibrated material model provides the
possibility of modelling the response of underground structures in various soil types subjected to
ground shock.
Keywords: Ground shock; Peak pressure; LS-DYNA; Numerical simulations
208
POSTER ABSTRACTS
Simplified Analysis for Large Deflections of Metal Sandwich Beams with

Lattice Cores under Impulsive Loading
Qinghua Qin* and L.H. Poh
State Key Laboratory for Strength and Vibration of Mechanical Structures, Department of
Engineering Mechanics, Xi’an Jiaotong University, China
Department of Civil and Environmental Engineering, National University of Singapore,
Singapore
*
Sandwich structures have been paid more attention due to their excellent blast resistance. It is
helpful for the designs and applications of the blast-resistant structures to explore the advantages of the
sandwich structures over the conventional monolithic solid structures with the same weight. In this
paper, a simplified analytical procedure is proposed to predict the dynamic response of large deflections
of metal sandwich beams with lattice cores under impulsive loading. The interaction between the plastic
bending and stretching in deforming regions is approximated by straight line in deformation region. The
dynamic response is decoupled into two phases and the closed-form solutions are obtained. Further,
effect of the bending stiffness is neglected and the metal sandwich beam is assumed to be a plastic
string. A closed-form solution is given then. The results show that the predictions of simplified
analytical models are in good agreement with the finite element (FE) results for the deflections and
structural response time of metal sandwich beams with lattice cores.
Keywords: Blast dynamic sandwich
209
AUTHORS’ INDEX
A Choon Keat Ang ........................................... 86

Chor Boon Ng ............................... 93, 135, 168
A.C. van den Berg ....................................... 143 Chua Hiong Suan .......................................... 94
Abraham Christian ...................................... 126 Chul-Gi Suk ................................................ 154
Adam Wisniewski ......................................... 83 Chunping Xiang ...................................... 75, 76
Ahmad Serjouei .......................................... 207
Alex Remennikov ......................................... 99 D
Alexander Siefert ........................................ 142
Ali Amini ...................................................... 87 D. X. Xiong .................................................. 92
Anand Sarma................................................. 95 D.W. Shu ....................................................... 98
Anh Tuan Pham..................................... 29, 172 Damith Mohotti ............................................ 99
Arno Klomfass ............................................ 144 Dan-Bi Kim ................................................ 173
Asif Usmani ........................................ 196, 194 David Cormie ............................................. 107
Atsuhisa Ogawa ............................................ 79 David Morin ............................................... 124
David Smith ................................................ 107
B David Stevens ............................................... 85
David Thambiratnam .................................. 132
B Janzon ...................................................... 145 Dawid Pacek ................................................. 83
Bamber R. K. Blackman ............................. 107 Deyang Kong .............................................. 182
Barry L. Bingham ....................................... 151 Diyuan Li .................................................... 160
Beng Chye, Vincent Tan ....................... 80, 168 Dongfack Guepi Clovis Jamot .................... 141
Bin-Chao Li ................................................ 195 Dong-WooShin ........................................... 195
Bing Li ........................................................ 171
Bo Yang ....................................................... 127 E
Bojan Zlender...................................... 122, 196
Brian Uy ........................................................ 99 Earl Maher .................................................. 180
Edgar Mataradze ......................................... 197
C Eng Choon Leong ............................... 108, 131
En-Hua Yang ....................................... 109, 169
C. G. Tay ..................................................... 153 Eu Chang, Leonard Heng ............................. 95
C.J. Shen ..................................................... 119 Eunsun Jo .................................................... 100
Carrie Davis .................................................. 85
Chan Ghee Koh ........................................... 153 F
Chao Hang .................................................... 74
Chee Leong Ho ........................................... 194 Fangrui Zhang............................................. 199
Chen Caihua ................................................ 133 Feng Xi ....................................................... 182
Cheng Xiong ............................................... 158 Francois Fayad ............................................ 140
Chengqing Wu .................................... 179, 199 Fritz-Otto Henkel ........................................ 142
Chenhui Wei ................................................ 161 Fumiya Togashi .................................... 87, 197
Cheol-Min Cho ........................................... 204
Cheolwoo Park ............................................ 201 G
Chi King Lee ....................... 121, 128, 175, 178
Chin Wai Aw ............................................... 148 G Meyer ...................................................... 145
210
AUTHORS’ INDEX
G. Lu ..................................................... 98, 119 Jian Jiang ............................................ 196, 194

Gai-Hui Zhang ............................................ 195 Jian Weng.................................................... 128
Gao-Feng Zhao ........................................... 159 JianChun Li................................................. 162
Ge Kang ...................................................... 129 Jiang Wenqiang ........................................... 166
Gui-Ping Zhao............................................. 195 Jianxun Zhang............................................... 75
Guo-Qiang Li............................. 117, 196, 194 Jian-Yun Sun ............................................... 117
Jichun Zhang............................................... 118
H Ji-HunChoi.................................................. 203
Jing Yan Kong............................................... 86
H.J. Li ......................................................... 119
Jinglei Yang ................................................ 109
Haibo Li ...................................................... 162
Jin-Young Lee ....................................... 81, 173
Hejia Wang .................................................. 206
Jiong Wei .................................................... 161
Heng Chye Chng ......................................... 137
Jiri Pachman ....................................... 101, 177
Heow Pueh Lee ........................................... 168
Jiri Stoller ................................................... 177
Hiong Suan Chua .......................................... 95
Jishuo Wang ................................................ 138
Hironori Ueno ............................................... 79
Joanna Kozłowska ........................................ 83
Hiroshi Andoh ............................................... 78
John E. Crawford .......................................... 77
Hoki Ban ..................................................... 201
John P. Dear ................................................ 107
Hongyuan Zhou .................................. 125, 149
Jong Yil Park....................................... 100, 141
Hoon Park ................................................... 154
Joo Ha Lee .................................................. 198
Hui Wei Yeo .......................................... 88, 135
José L. Torero ............................................... 57
Hyon-Soo Kim ............................................ 154
Josef Fladr................................................... 177
Hyung-Jin Choi ............................................. 77
Joseph D. Baum .................................... 87, 197
Hyung-Seop Shin .......................................... 96
Jun Li .......................................................... 179
Hyung-Sik Yang .......................................... 154
Jun Liu .......................................................... 69
Hyung-Won Kim ........................................... 96
Jung Hun Kee ............................................. 100
Hyun-Hoon Choi........................................... 81
Junyu Qian .................................................. 160
Hyunmyung Cho ......................................... 201
K
I
Kai Fischer .................................................. 144
Idapalapati Sridhar ...................................... 207
Kai Qian ...................................................... 171
Kang Hai Tan 29, 127, 128, 169, 172, 175,
J 178, 205
Kang Kok Wei .............................................. 94
J. Verreault .......................................... 143, 146
Kang Wei Lim............................................. 137
J. Weerheijm........................................ 134, 143
Kang-Bo Yuan .............................................. 82
J. Y. Richard Liew ........................... 92, 97, 153
Katsunori Fukui ............................................ 78
Jae-Ho Shin ................................................. 195
Ke Wei, Joel Ong .................................. 88, 168
Jaeyoon Kim ............................................... 180
Keita Iwano................................................... 78
Jang-Ho Jay Kim......................... 195, 203, 204
Kenneth Morrill .......................................... 150
Jaqueline Bell ................................................ 87
Khim Chye, Gary Ong ........................ 126, 176
Jia Han Chew .............................................. 131
Kian Wee Ng ............................................... 169
211
AUTHORS’ INDEX
Kinya Kuriki ................................................. 78 Ming Shu Tong ........................................... 135

Ki-Yeon Kwon .............................................. 81 Mingshi Wang ....................................... 76, 200
Knut Gaarder Rakvåg ................................... 94 Minkwan Ju ................................................ 201
Kun Du ........................................................ 158 Moshe Ravid ............................................... 151
Kwang Mo Lim ........................................... 198
Kwong Ming Tse ......................................... 168 N
Kyoung-Hwan Min ..................................... 173
Kyung-Oh Bae .............................................. 96 Na-HyunYi.................................................. 203
Namyo Salim Lim .............................. 128, 178
L Nimi Shapira ............................................... 151
L. Pereira ..................................................... 134 O

L.H. Poh ...................................................... 209
L.J. Sluys..................................................... 134 Oi Yin, Karen Chong .................... 88, 135, 168
Laddu Bhagya Jayasinghe........................... 132 Orlando A. Soto .................................... 87, 197
Lado Riannevo Chandra ..................... 126, 176
Lee-Ju Park ................................................... 96 P
Lei Dai .......................................................... 74
Paolo Del Linz ............................................ 107
Leilei Niu .................................................... 161
Paul A. Hooper ........................................... 107
Li Yulong..................................................... 133
Paweł Zochowski .......................................... 83
Long Bin Tan .............................................. 168
Peijie Yin .................................................... 159
Lukasz Wierzbicki ........................................ 83
Peiqi Ren..................................................... 174
Luke Pascoe ................................................ 107
Pengfei Wang .............................................. 109
Pengtao Wu ................................................. 179
M
Pengwan Chen ............................................ 129
M Kumm ..................................................... 145 Peter J Hoad ................................................ 148
M. D. Goel .................................................. 120 Peter W Weber ............................................ 110
M.J. van der Jagt-Deutekom ....................... 146 Phillip L. Gould ............................................ 46
M.M. van der Voort ............................. 143, 146 Piotr Sielicki ................................................. 84
Ma. Bavan ................................................... 167 Primoz Jelusic ..................................... 122, 196
Magnus Langseth .................................. 67, 124 Pu Xue ........................................................ 152
Marcin Leonowicz ........................................ 83
Marek Foglar....................................... 101, 177 Q
Martin Kovar............................................... 177
Qian Shi ...................................................... 114
Martin Künzel ............................................. 101
Qiang Pan ................................................... 118
Masuhiro Beppu ............................................ 79
Qinghua Qin ........... 75, 76, 114, 116, 200, 209
Matt Barsotti ................................................. 85
Miao Li ....................................................... 202
R
Michael M. Swisdak, Jr. .............................. 147
Michael Netherton ........................................ 99 R. Lohani .................................................... 120
Michael Stone ............................................. 180 R.M.M. van Wees ....................................... 146
Mikheil Chikhradze .................................... 197 Radek Hajek ............................................... 177
212
AUTHORS’ INDEX
Rainald Löhner...................................... 87, 197 T. Y. Wang ..................................................... 92

Raymond H. Bennett ................................... 151 T.Ch. Madhavi ............................................ 198
Rene Richter................................................ 123 Tae-Kyun Kim ............................................ 204
Rickard Forsén ............................................ 181 Takaaki Inuzuka ............................................ 78
Robert A. Frank ............................................. 87 Takayuki Ochikawa ...................................... 78
Robin French............................................... 180 Tamir Blass ................................................. 151
Rojana Pornprasertsuk ................................ 115 Tao Suo ......................................................... 74
Russell Norris.............................................. 150 Tao Wang .................................................... 200
Ryan McCuiston ......................................... 115 Teck Chee, Anthony Goh ............................ 132
Teng Sheng Peh .................................. 135, 136
S Tengjiao Cheng ........................................... 160
Theodor Krauthammer................................ 130
S. Pradyumna .............................................. 120
Theodore Krauthammer ................ 18, 180, 197
S.D. Brouwer .............................................. 146
Thomas Hartmann ...................................... 123
Samuel Rigby ................................................ 84
Tianyi Liu ................................................... 117
Sang-Ho Cho............................................... 154
Tim Brewer ................................................... 77
Sang-Won Lee ..................................... 203, 204
Truong-Thang Nguyen ............................... 205
Sarma Anand ................................................. 80
Tuan Ngo ...................................................... 99
Satadru Das Adhikary ......................... 126, 176
Tyler J. Ross ............................................... 147
Serdar Astarlioglu ....................................... 140
Seung Jae Lee ............................................. 100
V
Seung-Jai Choi ............................................ 195
Seungwon Kim............................................ 201 V. A. Matsagar ............................................ 120
Shahrizan Bin Baharom .............................. 167
Shalva Marjanishvili ........................... 140, 197 W
Shaobo Chai ................................................ 162
Shao-Bo Kang ............................................. 169 Wancheng Zhu ............................................ 161
Shengrui Lan ......................................... 77, 150 Wang Yonghu .............................................. 133
Shu Herng Ng ............................................. 137 Wee Teik Khoo ........................................... 137
Shuangjun Shen .......................................... 202 Wei Ping Cheng ............................................ 86
Siau Chen Chian ........................................... 80 Wei Zhang ................................................... 202
Siew Chin Lee ............................................... 97 Weifeng Yuan .............................................. 138
Simon K. Clubley........................................ 183 Wei-Guo Guo ................................................ 82
Siti Aminah Osman ..................................... 167 Weilong Ai .................................................. 114
Song Xuliang .............................................. 170 Weizhang Liao .................................... 202, 206
Stefan Greulich ........................................... 123 Werner Riedel ............................................. 144
Su Chern Tan ............................................... 148 Wijesundara M.G. Lakshitha ...................... 183
Sung-Wook Kim ......................................... 141
Suwen Chen ................................................ 117 X
X.M. Xiang ................................................... 98

T
Xiancheng Yu................................................ 69
T. J. Wang........................ 75, 76, 114, 116, 200 Xianfeng Zhang .......................................... 207
213
AUTHORS’ INDEX
Xiangyu Li .................................................... 98 Yonghui Wang .............................................. 97

Xiaojuan Wang .................................... 125, 149 Yongqing Chen ........................................... 127
Xiao-Ling Zhao ........................................... 199 Youcai Wu..................................................... 77
Xiaoping Zhang .......................................... 129 Youjun Ning................................................ 129
Xiaoshan Lin ............................................... 121 Young Hak Lee ........................................... 100
Xiaozhao Li................................................. 106 Young-Hun Ko ............................................ 154
Xibing Li ..................................................... 160 Young-Soo Yoon ................................... 81, 173
Xin Cai ........................................................ 158 Yu Jun ......................................................... 170
Xin Zhang ................................................... 109 Yu Su........................................................... 179
Xinkuan Zou ............................................... 118 Yu Zhang..................................................... 182
Xinzheng Lu................................................ 174 Yuan Zhao ................................................... 158
Xu Long ...................................................... 175 Yueqiang Li................................................. 182
Xuehui Yu ................................................... 116 Yukolthorn Sriloy ....................................... 115
Xue-Ming Tan ............................................... 82 Yulong Li ...................................................... 69
Yulong Zhou ............................................... 174
Y
Z
Y.X. Zhang .................................................. 121
Ya Dong Lin ................................................ 136 Z. Chen ....................................................... 152
Yadong Lin .................................................... 93 Z.H. Wang ................................................... 119
Yang Wang .................................................... 74 Ze Liang Yu ................................................ 152
Yang Wengang ............................................ 166 Zhen Cheng................................................. 161
Yew Soon Ng .............................................. 194 Zhengjin Wang ................................ 75, 76, 200
Yi Li ............................................................ 174 Zhimin Tian ........................................ 202, 206
Yi Wang....................................................... 107 Zhiye Zhao .......................................... 125, 149
Ying-Kuan Tsai ........................................... 130 Zhongbin Tang ........................................ 69, 74
Yip Choy Lau .............................................. 137 Zhongxian Liu ............................................ 179
Yong Kiang Tan .......................................... 194 Zhuoyuan Cheng......................................... 108
Yong Soo Yoon ........................................... 198 Zhushan Shao ............................................. 106
Zilong Zhou ................................................ 158
Zvi Asaf ...................................................... 151
214
EXHIBITION FLOORPLAN
S/N Organisation Name Booth No.

1 Advance Pinnacle Technologies Pte Ltd 1
2 Dyna Forming Engineering Technology Sdn Bhd 2
3 Worley Parsons Pte Ltd 3
4 Prostruct Consulting Pte Ltd 4
5 Advance Contech (Singapore) Pte Ltd 5
6 Nanyang Technological University – Nanyang Centre for 6
Undergound Space
7 Nanyang Technological University - Protective Technology 7
Research Centre
214
LIST OF EXHIBITORS
Advance Contech (Singapore) Pte Ltd

192 Waterloo Street, #05-01, Skyline Building
Singapore 187966
http://www.advance-contech.com/
Advance Pinnacle Technologies Pte Ltd

627A Aljunied Road, #06-02 Biztech Centre
Singapore 389842
http://www.apt.com.sg/
Dyna Forming Engineering Technology Sdn Bhd

534A, Jalan Seremban Tiga 15, Seremban 3,
Seremban 70300, Negeri Sembilan, Malaysia
http://www.dfe-tech.com/
215
LIST OF EXHIBITORS
Prostruct Consulting Pte Ltd

Block 165, Bukit Merah Central, Unit #04-3661
Singapore 150165
http://www.prostruct.com.sg/
Nanyang Technological University – Nanyang Centre for Underground

Space
50 Nanyang Avenue, N1-Blc-27
Singapore 639798
http://www.ptrc.ntu.edu.sg/Pages/default.aspx
Nanyang Technological University - Protective Technology Research

Centre
50 Nanyang Avenue, N1-B3-03
Singapore 639798
http://www.ptrc.ntu.edu.sg/Pages/default.aspx
Worley Parsons Pte Ltd

111 Somerset Road, #12-05 Tripleone Somerset
Singapore 238164
http://www.worleyparsons.com/
216
SHAPING
THE BUILT ENVIRONMENT

DAPS2015 Abstracts

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

DAPS2015 Abstracts

Caricato da

Copyright:

Formati disponibili

Table of Contents

ZHAO Zhiye Leonard HENG Eu Chang

International Advisory Committee

International Scientific Committee

Conference Delegates and Entitlements

Chui Huay Lim Club

Time Tuesday, 19th May 2015

0800 - 0830 Registration @ Venus Ballroom Foyer

0830 - 0900 Opening Ceremony @ Venus Ballroom I & II

Keynote – Recent Observations on Design and Analysis of Protective Structures

Keynote – Numerical Studies on Reinforced Concrete Sub-Structures under

1020 - 1050 Teabreak & Poster Viewing @ Jupiter Ballroom

Material Behaviour under Blast Numerical Modelling and Simulation I

A Micromechanical Method to Predict

Reaction Force and Delamination Design and Optimization of Laminated

Finite Element Analysis of Dynamic

Strain Rate Effects of Compressive Structural Optimization of K4-rating

Prediction of Penetration Depth of

Lunch @ Mercury Ballroom

Performance Based Numerical and Finite Element Analysis of

An Analytical Model for

Quantitative Risk Analysis

1510 - 1540 Teabreak & Poster Viewing @ Jupiter Ballroom

Time Wednesday, 20th May 2015

0830 - 0900 Registration @ Venus Ballroom Foyer

1020 - 1050 Teabreak & Poster Viewing @ Jupiter Ballroom

Structural Behaviour of Precast Beam-

Lunch @ Mercury Ballroom

Time Wednesday, 20th May 2015

Extended Beam Element

The Behaviour of Steel and Observations from Blast-

1510 - 1540 Teabreak & Poster Viewing @ Jupiter Ballroom

Time Wednesday, 20th May 2015

Design, Analysis and Numerical Modelling of Experimental Study of

The Use of Fiber

Nested Tube System Effect of Seismic Barrier Experimental Investigation

1900 Conference Banquet

Time Thursday, 21st May 2015

0830 - 0900 Registration @ Venus Ballroom Foyer

Keynote - Structures Subjected to Extreme Loading Conditions. Do We Need New

1020 - 1050 Teabreak & Poster Viewing @ Jupiter Ballroom

Structural Behaviour and Damage IV Numerical Modelling and Simulation V

Modelling Dynamic Tensile Failure of

Experimental Frequency Domain

Response to RC Slabs subjected to Assessment of Fast-Running Algorithms

End Restraint Effect on the Dynamic

Lunch @ Mercury Ballroom

Time Thursday, 21st May 2015

Application of Cellular Automaton in Dynamic Properties of Water Saturated

Using Performance-based Fire Numerical Study of Dynamic Fluid Flow in

Analytical Evaluation of Bi-

Design of Wireless System for the

1510 - 1530 Tall Buildings in Post Earth Quake Fire

1530 – 1600 Teabreak & Poster Viewing @ Jupiter Ballroom

1600 - 1630 Closing Ceremony & Best Paper Award Presentation

Tuesday, 19th May 2015

Recent Observations on Design and Analysis of Protective Structures

This equation could represent either a single-degree-of-freedom (SDOF) approach or an

2. Observations Related to Explosively-Induced Loads

2.1 Combined blast and fragment effects

2.2 Blast load suppression

2.2.1 Thermodynamic suppression

(a) Test set up

(b) overpressure histories, 1 - without mist, 2 - with water mist