Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Table of Content
Table of Content
Table of Content ....................................................................................... 1 Welcome message .................................................................................... 3 en!e ....................................................................................................... " International Workshop on Renewable Energy 2013 #ea$ership ............... % &ession 1 ................................................................................................... ' Implementation of renewable energy in power system ........................... '
Insulation Condition Monitoring and Diagnostics in Grid-connected Renewable Energy Systems..... 7 Integration of Renewable Energy Sources to Small Power Systems ................................................. ! " #wo-Stage Model Calculated Distribution System Planning Integrated Distribution Generator ... $ P%oto&oltaic 'ower system in (ietnam and )a'an* 'otential and de&elo'ment .............................. + Grid connected P( system facing &oltage sags , solutions to "&oid -nwanted disconnection ........ !. Electricity Su''ly to a /ocal0Isolated "rea by Means of Renewable Energy ..................................... !! "nalysis of (oltage Sags and Protection Coordination in Distribution Systems wit% Sensiti&e E1ui'ment ......................................................................................................................................... $
&ession 2 ................................................................................................. '( Technical) economic an$ policy iss!es for renewable energy .................. '(
Renewable Energy De&elo'ment in (ietnam .................................................................................... 7. #ec%nical Issues for 2ew and Renewable Energy De&elo'ment in (ietnam ..................................... 3. Medium-term and S%ort-term Electricity Demand 4orecasting ........................................................ 3$ Reconstruction of Syrian Electric Power Infrastructures by Renewable Energy ............................. 557 Current Problem 4aced in (ietnam "ssociated wit% t%e Integration of Small 6ydro and 7ind Energy into (ietnamese Electricity Grid ...................................................................................................... 5 3 "''lication of Renewable Energy to an -nder De&elo'ing Country8 9ambia ................................. 5$$ "n Introduction to 7ind Power in (ietnam* #ec%nology8 Grid Im'acts and Current Situation ...... 5:!
Table of Content
Design and ;'timi>ation of a Micro-cogeneration System -sing a ?Double Effect? Stirling Engine and a #ubular /inear Induction Generator ...................................................................................... ==3 /ig%tning Induced ;&er&oltage in #%e Control System of " 7ind #urbine ..................................... = ! Im'lementation Su'er&isory Controller for 6ybrid 7ind Microgrid System -sing "da'ti&e 2eural Mimo Model .................................................................................................................................... =+3 Modeling of Corona Disc%arge and Its "''lication to /ig%tning Electromagnetic Pulse Com'utations ......................................................................................................................................................... =$$ 2ew "lgorit%ms for Im'ro&ing t%e Reliability of 7ireless Sensor 2etwor@ in Renewable Energy Systems ............................................................................................................................................ =75
&ession " ............................................................................................... 2+% Win$) solar an$ ti$al energy ................................................................. 2+%
Direct 7a&e Energy Con&erters. Case of SE"RE( 'ro<ect ............................................................... =:$ ;'timal ;'eration of Isolated 7ind- diesel Power System - a Case Study ..................................... 5 " Pro'osal of MPP# Control for P( System under Partial S%aded Conditions ................................ .
Solution for /ocal Isolated Grid wit% 6ybrid System Including 7ind #urbin Interconnection ........ +3 ;'timi>ation "''roac%es for Si>ing and Placing Distributed Generators in a 2etwor@ System ..... $. Grid Integration Study of 7ind Power in Ain% #%uan Pro&ince....................................................... 3.
Welcome message
Welcome message
It is a great &leasure and $onor to welcome you to t$e 1+*< International 0or=s$o& on Renewable Energy ?I0RE 1+*<@3 being $eld in /anoi 2ni%ersity of Science and "ec$nology3 from 6ctober 1 to <3 1+*<. Renewable energy is an emerging t$eme for &ower system engineers in t$e last decades3 es&ecially after .u=us$ima disaster. Significant Auantities of renewable energy in &ower systems reAuire not only great deals of efforts in tec$nical wor=s3 but also in economic and &olitical matters. "$is wor=s$o& brings toget$er multidisci&linary eB&erts from academia3 industry and go%ernment institutions to discuss t$e c$allenges of t$e day of renewable energy in &ower systems3 and &ro%ide a &latform for s$aring of ideas and eB&eriences as well as for strengt$ening researc$ collaboration. I would li=e to ac=nowledge t$e !RC2S &ro7ect entitled CCollaboration Ile de .ranceDMidi-PyrEnEes a%ec le 'ietnam en rec$erc$e et formation F for t$eir original ideas of organi:ing t$is wor=s$o& as well as for t$eir financial and tec$nical su&&orts. I would also li=e to t$an= all members of local organi:ing committee for t$eir diligently %olunteer efforts3 wit$out w$ic$ t$is wor=s$o& would not be &ossible. S&ecial t$an=s to t$e sc$ool of Electrical Engineering of /2S" for t$eir su&&orts to $ost t$e Gala dinner. 0e $o&e you en7oy your stay $ere in /anoi and t$at your eB&ectation will be fully reali:ed.
!ssociate Professor "o& "ran 'an C$airman of I0RE 1+*< 'ice-&resident of /anoi 2ni%ersity of Science and "ec$nology
Venue
Venue
"a Guang >uu 5ibrary3 /anoi 2ni%ersity of Science and "ec$nology * Dai Co 'iet street3 /anoi
.rom *ate 1 on Dai Co 'iet road 0al= t$roug$ t$e C* building and t$en t$e main sAuare. .ollow t$e small &at$ to t$e left and reac$ t$e "a Guang >uu 5ibrary by going across t$e main &at$ of /2S" cam&us .rom *ate 2 on "ran Dai -g$ia street Go straig$t about 1++m and t$e "a Guang >uu 5ibrary is on your left
Technical committee
-ami$ 1en 0hme$3 #cole Normale Suprieure de Cachan, France iet 2g!yen+3!an+-oang3 Hanoi University of Science and Technology, Vietnam 4hanh 1ach+5!oc3 Hanoi University of Science and Technology, Vietnam 1ernar$ 6o!rnet3 #cole Normale Suprieure de Cachan, France Toan /h!ng, University of Ne$ South %ales, &ustralia T!an 2g!yen+0nh3 nstitute of Energy, Vietnam &on Tran+Thanh3 Electric !o$er University, Vietnam T! !+/han3 Vietnam National University, Ho Chi 'inh City, Vietnam
Session 1
#ession 1
C$airs)
Prof. "ran 'an "o&3 Hanoi University of Science and Technology Prof. !=i$iro !metani3 *oshisha University
Session 1
4eynote #ect!re
Abstract - "$e dri%e for clean energy and sustainability $as led to t$e emerging trend of integration of Renewable Energy ?RE@ systems into eBisting electricity grids. 0orldwide trends s$ow t$e &ro&ortion of grid-connected generation from renewable energy ?$ydro3 solar3 wind3 etc@ is steadily increasing. "$e im&act of $ig$ &enetration of distributed RE into t$e electricity grid &resents new c$allenges to t$e reliable o&eration of t$e infrastructure3 e.g. cables3 transformers3 etc. "$is &resentation gi%es an o%er%iew of %arious tec$nical issues im&osed on t$e insulation systems of &ower &lant eAui&ment. 6f &articular concern are t$e increased ris= to t$e insulation caused by $ig$-freAuency switc$ing transients and o%er%oltages generated by &ower electronic interfacing. "$e $ig$er degree of generation intermittency will affect &lant o&eration3 t$ermal stress &attern3 and t$us insulation ageing. 6%er t$e years3 one of t$e =ey researc$ acti%ities at 2-S0 $as been t$e de%elo&ment of insulation diagnostic tec$niAues and on-line conditioning monitoring systems. Muc$ of t$is wor= can be utili:ed to assist in e%aluating t$e insulation &erformance under new o&erating en%ironments. "$is will be discussed in t$e tal=.
2 October, 2013
10
Australia generates ~1.5% of global greenhouse gas emissions. On per capita basis, one of world's largest polluters. About 24.4 tonnes of CO2e per person in a year (2012) About twice OECD average and > 4 times world average Sources of Australias emissions:
12
Source: Department of Climate Change and Energy Efficiency Australias Emission Projections 2010
with high wind resources (speeds >7.5m/s). Annual estimate of 273 TWh.
Biomass Energy: Australia is appraised potential of bioenergy resource is
receives annual average rainfall of <600 mm limited hydro power resource (Snowy Mountains, Tasmania)
Ocean Energy: tidal energy, wave energy and ocean thermal energy. Total
13
Annual average solar radiation (in MJ/m2) and currently installed solar power stations with capacity of >10 kW
Source: Australian Energy Resource Assessment
14
15
16
140 x 3MW turbines, currently the largest wind farm in the southern hemisphere. Cost AUD1 billion, construction took 2.5 years, operational since late Jan. 2013. Wind turbine output: 480-690V, convert by WTSU transformer to a collector voltage of 13.8-46kV. ~100km of underground 33kV cable and OH lines connecting turbines to the wind farm sub-station. 33kV/132kV wind farm sub-station with 2 x 280MVA parallel step-up transformers. 14km of 132kV OH lines to Tarrone sub-station where connection to 500kV line was made.
17
Australias per capita electricity consumption: ~22% > OECD average Capacity of grid-connected generation = 54.3GW Capacity of embedded and non-grid generation = 5.9GW
Total generation (2010-11) = 228,067 GWh Black coal: 50.7% Brown coal: 24.2% Natural gas: 15.5% Hydro: 7.0% Wind: 2.6% Oil and other: <0.1%
18
Steady state and transient stability analysis Load/generation coincidence (peak load and variability of source) Regulation requirements Integration with Automatic Generation Control (AGC) Incorporation of renewable resource forecasting Operating practice to enable high penetration
Voltage and VAR regulation Power Quality (Harmonics, Flicker, DC Injection) Unintentional islanding Protection design and coordination Equipment grounding Load and generation imbalance Energy storage and storage controls
19
good for insulation More repeated thermal cycling hot spots, PDs accelerate insulation ageing higher risk of insulation failure than other transformers More frequent operation of on-load tap changers (OLTC) Solid-state switching and non-sinusoidal waveforms harmonic voltage frequencies Requirement to ride-through transient events and faults exposure to large electro-dynamic forces from short circuits Large number of WTSU transformer failures need more robust design.
20
http://www.csanyigroup.com/the-unique-role-of-wind-turbine-step-up-wtsu-transformers
Transformers: Core loss (iron loss) Hysteresis loss Eddy current loss Winding loss (copper loss) Cables: Dielectric loss
n 2 P k1 fBm k2 f 2 Bm
P 2 fCV 2 tan
Skin effect: higher frequency reduced skin depth higher resistance higher Ohmic loss Increased power loss Lower efficiency Higher operating temperature Derating of equipment
21
Harmonics: increase real power loss insulation temperature rise accelerate ageing (Arrhenius law)
High-frequency switching spikes increase likelihood of partial discharges (PD). detection of PDs during HF transient is a challenge. HV-DC enabling technology for access to remote RES, e.g. offshore wind farms, Europe Multi-terminal DC Super Grids electric stress under DC very different from that under AC space charge temperature effect is very significant 22
Power system components: transformers, overhead transmission lines and cables, switchgear, rotating machines Choice of electrical insulation varies for particular application and voltage level
Wide range of electrical insulating materials Gases: air, Nitrogen, Hydrogen, SF6, SF6 mixture, vacuum, etc Liquids: mineral oils, synthetic hydro-carbons, etc Solids: resins, polymers, ceramic, etc Composites Electrical breakdown strength Real power loss: leakage current loss, dielectric loss (AC only), partial discharge (PD) Condition of insulation critical factor to equipment life
23
Condition monitoring
Asset Maintenance Strategies Corrective maintenance (CM): reaction only when failure occurs Time-based maintenance (TM): preventive maintenance in fixed time periods Condition-based maintenance (CBM): preventive maintenance depending on actual conditions Extend asset lifetimes through condition monitoring On-line condition monitoring: Smart Grid
24
Vales Point power station Transformer 25 blast and fire SMH 9Nov06
Insulation Resistance (IR) Polarization index (PI) Dielectric absorption ratio Dielectric dissipation factor (DDF) Dissolved gas analysis (DGA): Oil: H2, CH4, C2H2, C2H4, C2H6, CO, CO2 Cellulosic materials (paper): furans Partial discharge (PD)
Dielectric spectroscopy: Time domain: polarization depolarization current (PDC), RVM Frequency domain dielectric spectroscopy (FDS)
Over-voltage tests: Power frequency, very low frequency (VLF), HV-DC Impulse (lightning, switching)
26
Localized electrical discharge that only partially bridges insulation between conductors
Due to presence of small defects or design flaws which create a localized region of excessive electric stress that exceeds breakdown strength of the insulation
Example: electric stress in cables
MV cables (3.8/6.6 19/33kV XLPE single core screened and PVC sheathed)
27
28
Significant interference and background noise detection sensitivity is a challenge Measured (apparent) PD magnitude < true PD magnitude Extensive use of signal processing Cable joints, main cable insulation, cable terminations Different types, different ages, different operating history
PDM-I system: Suitable for good SNR situation Up to 12 channels multiplexed Fast quasi real-time monitoring
PDM-II system: Suitable of noisy locations Dual simultaneous channels Software-based PD extraction
31
Clamp sensor
32
Conclusion
Pressure of climate change and supply security dramatic increase in level of penetration of RES Of particular importance are wind and solar energy Significant changes to electricity grids operation Undesirable impacts by RES on electrical insulating systems Need for research on insulation systems behavior under the new operating environment, in particular under DC and combined DC/AC On-line insulation condition monitoring is an integral component for building smarter grids.
33
34
Session 1
Abstract - "$is &resentation deals wit$ c$allenges and countermeasures for integration of renewable energy sources to &ower systems3 es&ecially to small &ower systems suc$ as microgrids and small remote islands. C$allenges include demand and su&&ly control met$od considering uncertainty of generations and loads es&ecially forecasting &ower out&uts of renewable energy sources. !lso in t$is &a&er we deal wit$ &ower out&ut fluctuations mitigating tec$nologies for renewable energy sources. Efforts are &erformed as demonstration &ro7ects and as standardi:ation of microgrid tec$nologies. I li=e to consider $ow we could collaborate toget$er to de%elo& and commerciali:e integration of renewable energy sources to small &ower systems.
35
Session 1
Abstract - Recently3 t$e distribution system design and &lanning $as c$anged due to t$e im&acts of DG and electricity mar=et restructuring. "$erefore3 t$is &a&er &ro&oses a two stages model for o&timi:ing &lanning of distribution system wit$ t$e &resence of DG. "$e &ro&osed model can determine o&timal u&grading si:ing and timeframe of eAui&ment in distribution system. >esides3 o&timal dis&lacement3 si:ing3 tec$nology and installation &eriod of DG are also determined. "$e modelHs t$e ob7ecti%e function is minimum life cycle cost for t$e &lanning sc$eme. "$e constraints are used to guarantee t$e tec$nical and economic indicators of t$e system. "$e calculation &rogram is made in G!MS en%ironment. "$e feasibility and effecti%eness of t$e &ro&osed model are %erified by t$e result of a&&lying it to a test system. Keywords: Planning of Distribution System ?DS@3 Distributed Generator ?DG@
36
I. INTRODUCTION In the past decade, distribution system planning had major changed due to the impact of competitive electricity market, DG technological development and environmental pollutions. In particular, DG connecting directly to DS or directly supplying to customers is used as a popular planning approach. These sources normally use electric generating technologies such as gas turbines, combined heat and power, Fuel Cells, solar energy and wind energies. Therefore, the benefits of DG including reduction of transmission and distribution cost, power loss and enhancement of flexibility and reliability of DS, improvement of differential voltage at nodes as well as reduction of environmental pollution [1]. However, DG requires high investment, makes increasing the complexity in measurement and relay protection as well as operation of DS [2]. Besides, DG using renewable energy resources has the naturally variable power according to primary energy. Many planning models of the DG integrated distribution system are already been researched and proposed. The authors in [3] presented a long-term DS planning model in order to determine capacity, location and a new building investment process or to upgrade current equipments by using popular mathematical programming. The objectives of model are the minimum total of investment and operation costs of DG, the investing cost for feeder and substation transformers during planning period. The details of DG technology is not mentioned because of the assumption that the costing functions and effects of DG in DS planning are the same, but these are impossible in reality. Another model in [4] was proposed with
37
model is added in the second stage plan to more closely reflect the required investments and production schedules. A. The Mathematical Model of First Stage 1) Objective Function The objective function of proposed model is to minimize total life cycle cost of the investment project during calculation period as shown in equation(1). The total cost is calculated at base year with equation 1/ (1 + r)t and discount rate r.
J =
t =1 T
RNt =
+
k =1 i =1
K DG N DG
DG (tkh , k - TDG , k )
TDG ,k
DG CiDG , k .P i ,k ,t
(7)
" i N DG , k K DG , t T
2) The constraints a) Constraints for power flow The output power characteristics of each DG technology using renewable energy resources fluctuate by time of day and season in year so the power of DG is also determined by each hour, season and specially, each technology k of DG. Hence, a nonlinear power flow representation in (8) is used in this stage.
K DG k =1
"t T
Where, component CFt is upgrading costs of feeders for year t with fixed capital cost (CFF) and variable capital cost (CFC) as shown in equation(2).
CFt =
N
DG i ,k ,s ,t ,h N
+ Pi ,Ss ,t , h - PDi ,s ,t , h =
ij , t
i =1 j = i +1
L (C
ij
FF
.a ij.t + C FC .Fij ,t )
(2)
K DG k =1
Y
j =1
. U i, s ,t ,h . U j ,s ,t ,h .cos(qij ,t - d j , s ,t ,h - d i ,s ,t ,h ) (8)
S i , s ,t , h
"ij N , i j, t T
Substation transformers upgrading costs in year t with fixed capital cost (CSF) and variable capital cost (CSC) is presented in equation(3).
CSt = (C SF .g i ,t + C SC .DSiS ,t )
i =1 NS
DG i ,k , s ,t ,h
+Q
N
- QDi , s ,t ,h =
- Yij ,t . U i, s ,t ,h . U j ,s ,t ,h .sin(q ij ,t - d j ,s ,t , h - d i ,s ,t ,h )
j =1
(3)
"i, j N , k K DG , s SS , h H , t T Where, Pi,DG and QiDG are output power of DG k ,s ,t , h , k , s ,t , h introduction in(9).
DG DG Pi ,DG k , s ,t , h = P i , k , t .k k , s , h DG QiDG , k , s , t , h = cosj k .P i , k , s ,t , h
"i N S , t T
(9)
(4)
"i N S , t T , s SS , h H
The equation (5) is new investment costs in year t with technologies k of DG. Beside, electrical energy purchased cost from electricity market and costs for fuel, operation and maintenance of DG depending per technology k, operation season s and time h are shown in equation(5) and (6).
DG CDGt = CiDG , k .P i , k ,t i =1 k N DG K DG
b) Limit capacity constraints of DG These constraints allow computed DG capacity at nodes in limit of DG technology, and it ensures annually upgrading power corresponding to equipment parameters as shown in(10).
DG 0 Pi ,DG k ,t P i , k ,max DG DG Pi ,DG k ,t = P i , k ,t -1 + DP DG 0 QiDG , k ,t tan j k .P i , k ,t DG DG (10) QiDG , k ,t = Qi , k ,t -1 + DQ
(5)
"t 1, i N DG , k K DG , t T c) Upgrading section constraints of feeder Thermal limits are imposed to limit the loading of feeders and these limits take into consideration the new feeder investments. So, the feeders upgrading constraints and upgrading power satisfying equipment parameters are shown F in(11). A step increase of feeder capacity at year t ( DSij, is t) set when capacity value is equal or greater than limit capacity used at year t-1.
*F F Sijm,ax t ( S ij ,t -1 + DS ij,t ) F F DSij, t DS min .a ij,t
"i N DG , k K DG , t T
DG DG DG DG EDGt = Ds ( r P . k .P i , k , t , s , h + rQ . k .Qi , k ,t , s , h ) i =1 k =1 s =1 h =1 N DG K DG SS H
(6)
"i N DG , k K DG , t T , s SS , h H
The residual value of equipments at the end of the planning period is presented in equation(7) and it is usually evaluated basic on the current market conditions. Hence, the residual value is the present value and it is calculated at base year in objective function.
(11)
DS M .a ij,t
F ij,t
"t 1, ij N , t T
38
Then, feeder capacity needs to meet in order to supply power to the loads present in (12) and upgrading section is selected by equation(13) with current density J.
*F *F F Sij ,t = Sij ,t -1 + DSij, t *F Sij ,t
stage. Hence, equations of objective function are presented as . (16) and decision variable DG power is Pi ,DG k ,t
J2 =
t =1 NS
Fij,t
3U dm .J
.a ij ,t
d) Addition capacity constraints for substation These constraints allow to maximize the use of existing substations capacity and to satisfy upgrading power corresponding to equipment parameters. A substation capacity S addition step size (DSi, is used to set substation sizes as in t) equation(14) with the maximum and minimum allowable capacity which a substation can be upgraded.
ax S Sim ( Si*,tS-1 + DSi, ,t t)
DG DG DG DG + k s ( rP . k .P i ,k ,t , s , h + r Q. k .Qi , k ,t , s , h ) i =1 k =1 s =1 h =1
(16)
+ +
DS DS .g i,t
S i,t S i,t S min
(14)
DS M .g i,t "t 1, i NS , t T
e) Constraints of limited nodal voltage Technical requirement constraints of limited nodal voltage are given in equation(15). Voltages at substation nodes are assumed constantly.
(15)
The decision variables of model include real and binary variables so calculation results must be corrected by standard equipment in fact and used as parameters in second stage.
TABLE I. SETS AND INDICES No 1 2 3 4 5 6 7 8 9 Symbol N i, j NL NS NDG t, T h, H k,KDG s,SS Definition Set of buses in distribution system Bus (i, j N) Set of load buses in distribution system Set of substation buses in distribution system Set of DG buses in distribution system Planning year and overall planning period (t T) Hour and hours per day (h H) Technology and total technology of DG (k KDG) Season and total seasons in year (s SS)
(17)
"t 1, ij N , t T , s SS , h H
S SiS,t , s ,h S m ax.i , t
"t 1, i NS , t T , s SS , h H
(18)
c) Limit capacity constraints of DG The investment location and period of DG was determined from first stage so these constraints allow selected DG capacity according to new limits as(19).
* DG 0 Pi ,DG k ,t P max .i , k DG DG Pi ,DG k ,t = P i , k ,t -1 + DP DG 0 QiDG , k ,t tan j k .P i , k ,t DG DG (19) QiDG , k ,t = Qi , k ,t -1 + DQ
B. The Mathematical Model of Second Stage This stage takes the input parameters obtained from the first stage as addition capacity of substations, upgrading section of feeders, installation location and period of DG. Then, it determines the DG capacity within pre-defined bounds. 1) Objective Function The model has objective function similar first stage with S upgrading variables of feeders ( Fij,t ) and substations ( DSij, t) are replaced by equipment parameters obtained from the first
"t 1, i N DG , k K DG , t T The proposed comprehensive plan includes a MINLP model in first stage and NLP model in second stage. The calculation program is made in GAMS environment used MINOS solver [8] to find out an optimal solution. Sets, indices, variables, parameters and symbol of model are presented in table I, table II and table III.
39
TABLE II. PARAMETERS No 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Symbol r CFF CFC Li,j Yi,j,t, qi,j,t CSF CSC Definition Discount rate (%) Fixed capital cost of Feeder ($/km) Variable capital cost of Feeder ($/km.mm2) Length of Feeder (km) Magnitude and Angles of admittance matrix element (pu) Fixed capital cost of Substation ($/Substation) Variable capital cost of Substation ($/MVA) New investment cost for DG i, technology k ($/M)
III.
A. Diagram and Parameters of distribution system The 7-bus and 22kV voltage radial diagram is investigated in this research as figure 1 and is connected to 110kV substation. The total active power and reactive power at the base year are 6465kW and 5091kVAR, respectively.
1 5 2 6 3 4
DG i ,k
r hPS
QS rh DG rP .h
DG rQ .h
Active power purchased cost from market ($/kWh) Reactive power purchased cost from market ($/kVAh) O&M cost and Fuel cost of DG for active energy ($/kWh) O&M and Fuel cost of DG for reactive energy ($/kVAh) Active power demand at bus (kW) Reactive power demand at bus (kVAr) Maximum power limit of DG i, technology k (MW) Power factor of DG with technology k New maximum power limit of DG in second stage (MW) Standard section of Feeder in planning year t (mm ) Maximum capacity need upgrading of Feeder (MVA) Capacity ramp-up limit for Feeder (MVA) Maximum capacity limit of standard Feeder (MVA) Maximum capacity need upgrading of Substation (MVA) Capacity ramp-up limit for Substation (MVA) Maximum capacity limit of standard Substation in planning year t (MVA) Current density at thermal limit (A/mm2) Big number used maximum limit of variables in MIP and MINLP models Maximum voltage limit at bus (pu) Minimum voltage limit at bus (pu) Active power ramp-up limit for DG (MW) Reactive power ramp-up limit for DG (MVAr) Output power factor of DG with technology k Variation factor of the price of electricity Total day per season TABLE III. VARIABLES No 1 2
2
Substation
PDi,s,t,h QDi,s,t,h
DG max .i , k
Cosjk
DG Pm*ax .i , k * ij, t *F ij, t
B. Assumptions in analyis This research utilizes some economic and technical assumptions for the ease of computation: Planning period is 10 years and annual developing rate of load demand is constant, 10% per year The constructing cost of 110kV substation including fixed costs and variable costs is 0.2M$ and 0.05M$/MVA, respectively [5]. Similarly, the upgrading costs of 22kV feeders consist of 0.15M$/km and 0.001M$/MVA.km The effects of DG technology are represented by investment, operation and fuel costs. Two DG technologies, photovoltaic (PV) and small hydro sources, are used in this research with the corresponding capital costs to be 5.0M$/MW and 1.5M$/MW. Average O&M costs depend on used technology and the life of DG such as table IV. The assumption life of feeder is 20 year.
TABLE IV. AVERAGE O&M COSTS AND LIFESPAN OF DG Technology PV Small hydro Average O&M costs Active power Reactive power ($/kWh) ($/kVAr) 5 0 5 1 Lifespan (years) 20 30
F S S S
S DSmin S Smax.i, t
kkDG ,s ,h
kP DS
No 1 2 3 4 5 6 7 8 9 10 11 12
Symbol
Definition Upgrading section of Feeder (mm2) Addition capacity for Substation (MVA) New investment capacity of DG (MW) Active power purchased from electricity market (kW) Reactive power purchased from electricity market (kVAr) Addition capacity of Feeder (MVA) Active output power of DG (kW) Reactive output power of DG (kVAr) Voltage for bus (pu) Voltage angle at bus (pu) Binary variable on feeder upgrade decision (1/0) Binary variable on feeder upgrade decision (1/0)
Fij,t
S DSi, t
Pi ,DG k ,t P
S i ,s ,t ,h S Qi, s ,t ,h
Energy prices purchasing from electricity market through substations are specified in terms of the three time blocks of peak, intermediate and base price as table V.
TABLE V. ENERGY PRICES PURCHASE FROM ELECTRICITY MARKET Base 36.35 1.82 Intermediate 58.20 2.91 Peak 105.95 5.3 Time block Energy $/MWh price $/MVArh
DS P Q
F ij,t
DG i, k , s , t , h DG i, k , s , t , h
Ui,s,t,h di,s,t,h
a ij,t g i ,t
PV is manufactured in compact modules occupying small spaces and time to install is short. Hence, installing areas at load locations are not limited. In contrast, the location and rated capacity of small hydro are predefined dependence on primary energy source and assumption that they can be built in the 4, 6 nodes.
40
Areas of upgrading of substation transformers and feeders are not limited Constraint of limited load nodes voltage changes from 0.9pu to 1.1pu, and it should be 1.05pu at substation node Decided variables in the model are continuous in order to reduce the complexity of the model. Hence, they should be rounded to match real equipments.
economic and technical efficiencies are gained. In contrast, the PV is not selected due to a very high investment cost.
TABLE VI. FEEDERS UPGRADING DECISIONS Feeder Feeder section upgrading in each year (mm2) 1 2 3 4 5 Case A 6 7 8 9 70 10
C. The output power characteristics of DG The output power of PV depends on the intensity of solar radiation and its performance. The power of 1MWp PV with 25% performance calculated basing on the given solar radiation intensity is presented as figure 2.
The out power of PV, MW 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 hour Summer Winter
50
Economic indices are compared between case B and case A as in table VIII. The case B holds a better economic index. Cost of DG investment and equipment upgrading (feeders and substation) are more expensive than those of case A about 2.17M$ due to a very high cost of DG investment. However, O&M and electric energy expenses have been decreased by 4.26M$ because of very low O&M expenses of small hydro. Therefore, the efficiency gets higher at final years of planning period. Total life cycle cost of case B is cheaper than these of case A by 3.47M$, equal to 17.2%.
TABLE VII. DG INVESTMENT DECIDED DG Bus technology PV Small hydro Total 2 7 4 6 DG capacity invested in each year (MW) 1 2 3 4 5 6 7 8 9 10
Small hydro power depends on the nature of the primary energy source. Therefore, the output power characteristics of 1MW small hydro are shown as figure 3.
The out power of sm all hydro, MW 1.2 1 0.8 0.6 0.4 0.2 0 1 2 3 4 5 Summer 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Winter hour
1 1 2.0MW
D. Analysis results and disscussions The feasibility of the proposed model and efficiency of DG are investigated in two cases. Case A: DG is not considered when calculating DS planning. Case B: DG is integrated in the researching model. The results of calculating showed that case A need to upgrade substation with a 10MVA capacity. In contrast, investment to upgrade substation in case B is deferred because of the load demand increasing in the future is provided by DG. Similarly, the case Bs feeders are not also upgraded during the planning period. In the case A, 2 feeders need to upgrade in the time from 8th year to 9th year as represented in table VI. Table VII presents optimal investment decisions of proposed planning model for DG. The total of investment capacity during planning time is 2.0MW equivalent to 30.9 percent of base years load demands. Small hydro investment focuses mainly on the first years of planning period and selected location of DG is far from substation so high
The technical indicators of DS are also improved when DG is integrated on DS planning. The power loss at maximizing load demand times is reduced 1.8% in 7th planning years so electric energy loss decreased 3210.0MWh during planning period. Total of electric energy purchased from market is also decreased 97,740.0MWh corresponding to 25,402.6tons are CO2 emission, which contributes to the decrease of environmental pollution.
TABLE VIII. ECONOMIC INDICES COMPARISON No Cost Total life cycle cost (M$) Feeder and Substation upgrading cost (M$) O&M and Electrical energy cost (M$) Investment DG cost (M$) Case Case Comparison A B B and A 20.21 16.74 -3.47 1.26 0.43 -0.83 -4.26 3.0 Note Total life cycle cots is reduced 17.2%
1 2 3
4
The voltage loss on the feeders reduces because of DG has reduced the transmission capacity from the substation to the load. Therefore, voltage profiles at the all bus are also improved during calculation time. In particular, load node having the biggest support is 4-bus. This bus voltage profile increased from 0.95pu (case A) to 0.97pu (case B) at 18th hour in 7th planning year.
41
IV. CONCLUSIONS Recently, the DS planning has been changed significantly by the impacts of DG and environmental policies. DG has many benefits for DS as enhancement of flexibility and reliability, bus voltage improvement, reduction of transmission cost and power loss as well as reduction of environmental pollution. However, the investment cost of DG is expensive and DG power that uses renewable energy resources is natural variability according to primary energy so the planning and operation calculation of DS will be more
APPENDIX A. DATA OF LOADS No 1 2 3 4 Bus 1 2 3 4 PD0 (kW) 743 1525 650 QD0 (kVAr) 458 1217 507 No 5 6 7 Bus 5 6 7 Total PD0 (kW) 1070 1830 647 6465 QD0 (kVAr) 873 1608 428 5091
difficult. Therefore, this study proposed a new two-stage optimized model that is integrated DG in DS planning. In this model, equipment sizing and timeframe required for upgrading equipment for DS well as select DG technologies with power variable constraints of DG can be determined. The objective function is minimizing total life cycle cost of the investment project. Calculation results showed that the proposed model is suitable in large DS planning calculations and the planning together with using DG provided better economic and technical indicators. REFERENCES
[1] [2] Thomas Ackermann, Goran Andersson, Lennart Soder, Distributed generation: a definition, Electric Power Systems Research 57, 2001 S. Wong, K. Bhattacharya and J.D.Fuller, Comprehensive framework for long-term distribution system planning, Proc. IEEE PES Annual General Meeting, Tampa, USA, 2007 Algarni, A.A.S.; Bhattacharya, K., A Novel Approach to Disco Planning in Electricity Markets: Mathematical Model, Power Systems Conference and Exposition, 2009. PSCE '09. IEEE/PES El-Khattam, W.; Hegazy, Y.; Salama, M., An integrated distributed generation optimization model for distribution system planning, Power Engineering Society General Meeting, IEEE, 2005 S. Wong, K. Bhattacharya1and J.D. Fuller, Electric power distribution system design and planning in a deregulated environment, IET Generation, Transmission & Distribution, 2009 V.V.Thang, D.Q.Thong, B.Q.Khanh, A New Model Applied to the Planning of Distribution Systems for Competitive Electricity Markets, The Fourth International Conference on Electric Utility Deregulation and Restructuring and Power Technologies (DRPT) 2011, Shandong, China, 2011 Su. H, Zhang. J, Liang. Z, Niu. S, Power Distribution Network Planning Optimization Based on Life Cycle Cost, 2010 China International Conference on Electricity Distribution, 13-16 Sept. 2010 Richard E. Rosenthal, GAMS - A User's Guide, GAMS Development Corporation, Washington, USA, 2010.
[3]
* Where: PD0 , QD0 - active and reactive power demand at bus in base year of planning period
[4] APPENDIX B. DATA OF FEEDER PARAMETERS No 1 2 3 4 5 6 Bus i - Bus j 1-2 2-3 3-4 1-5 5-6 2-7 Fij (mm2 ) 50 50 35 35 35 35 Smax.ij (MVA) 8 8 6.67 6.67 6.67 6.67 Lij (km) 2.3 2.2 3.3 3.5 1.7 1.2 Rf ij (W ) 1.362 1.302 2.551 2.706 1.314 0.928 Xfij (W ) 0.961 0.920 1.416 1.502 0.729 0.515 [5]
[6]
[7]
[8]
42
Session 1
*epartment of electric po$er systems, School of Electrical Engineering, Hanoi University of Science and Technology, 0. *ai Co Viet, Hanoi, Vietnam
Vietnam Ta1uchi Electric Co2, 3td, 3ot .4, *ai *ong Hoan Son ndustrial !ar,, 5ac Ninh !rovince, Vietnam
Abstract) Pur&ose I "$is &a&er see=s to e%aluate t$e solar energy &otential of 'ietnam. In com&are wit$ (a&an3 country $as t$e strongly de%elo&ment in t$e domain &$oto%oltaic &ower system3 we $a%e s$own t$e solar energy &otential of 'ietnam. DesignDmet$odologyDa&&roac$ I "$e a&&roac$ ta=es t$e researc$ of a real P' system in "abuc$i 'ietnam Com&any) Electricity &roduction3 economic analy:e .indings I "$e &a&er s$ows t$e efficient of t$e installation Grid-Connected P$oto%oltaic System in /anoi area. 6riginalityD%alue I E%aluation t$e efficient of t$e Grid-Connected P$oto%oltaic System in /anoi area is a &rereAuisite for in%est t$e furt$er researc$. Keywords: P' &ower system3 Grid-Connected P$oto%oltaic System in /anoi
43
Department of electric power systems, School of Electrical Engineering, Hanoi University of Science and Technology, 01 Dai Co Viet, Hanoi, Vietnam 2 Vietnam Tabuchi Electric Co., Ltd, Lot 13, Dai Dong Hoan Son Industrial Park, Bac Ninh Province, Vietnam Email: Tung.leduc-see@hust.edu.vn
ABSTRACT
Purpose This paper seeks to evaluate the solar energy potential of Vietnam. In compare with Japan, country has the strongly development in the domain photovoltaic power system, we have show the solar energy potential of Vietnam. Design/methodology/approach The approach takes the research of a real PV system in Tabuchi Vietnam Company: Electricity production, economic analyze Findings The paper shows the efficient of the installation Grid-Connected Photovoltaic System in Hanoi area. Originality/value Evaluation the efficient of the Grid-Connected Photovoltaic System in Hanoi area is a prerequisite for invest the further research. Keywords PV power system, Grid-Connected Photovoltaic System in Hanoi
1. INTRODUCTION
ENERGY production is a challenge of great importance for the coming years. Indeed, the energy requirements are increasing. Today, much of the world's energy is supplied from fossil sources. The consumption of these sources gives rise to gases emissions, greenhouse effect and thus to an increase in pollution. In addition that the fossil fuel resources are limited.
44
In the past 10 years, the use of renewable energy sources is seen as the solution to energy demands of the future. In recent years, there are an exponentially implementation of distributed generation (DG), in particular renewable energy (Photovoltaic, Wind). Photovoltaic technology has now become a major factor in the electricity sector globally. At least 110 TWh, 110 billion kWh in 2013 will be produced by photovoltaic systems already installed. If this represents approximately 0.5% of the electricity demand in the world, some countries have achieved significant percentages quickly (see Fig.1 and Table 1) [1]-[2]
Major sources of commercial energy in Vietnam are coal, petroleum, and hydropower. Significant number of households is using traditional solid fuels in residential sector for heating, light and cooking. Vietnam for instant is net exporter of energy due to its oil and coal resources. However, in the future, the energy problems in Vietnam will be not out of trajectory of the world. The research and the using solar PV technology are urgent issues. In this paper, the potential and development of PV system of Vietnam is analyze in compare with Japan-a power country of PV technology (Part 2). In part 3, a research of a Grid-connected PV system is presented.
Hanoi
1 0.5 0
Ho Chi Minh
Tokyo 1
Sapporo 2
Hiroshima 3
Hanoi 4
Danang 5
Hochiminh 6
Japan
Vietnam
Fig.2. Compare of Solar irradiation between Vietnam and Japan (North South)
45
The Fig.2 shows that the solar energy potential of Vietnam is higher when compare preliminary with Japan. As opposed to the potential, the PV development of Vietnam is a lot smaller. The total power of PV system of Vietnam is 4MWp, for instant. Whereas, the total power of PV system in Japan is 7GWp, and continue to rise sharply in the coming years (Fig.2). The reason for this development is due to the preferential policies of Japans government: Government enacted the Renewable Energy Law in 2011, Feed-in Tariff (FIT) program for renewable energy power generation facilities started from July 2012, Subsidy for measures to support introduction of residential PV systems (budget: 1,2 billion USD for multiple-years), Subsidy for introducing renewable energy power generation systems as part of restoration measures (budget: 316 million USD), 1096 local governments and municipalities were offering subsidy programs for residential PV systems [2]-[3]
kW
7,000,000
6,000,000
5,000,000
2,000,000
1,000,000
Thus, Vietnam is still missing the suitable policy of the government for development the PV system and renewable energies in general. In collaboration with Tabuchi Electric Company, the research of technical and economic has been invested. This is the content of part 3.
46
Let us consider a PV system connected to Grid (Fig.3). The installation power is 9.0 kWp, including five PV arrays (1.8 kW/array). The system is installed at Tabuchi Vietnam Company in Hanoi area. The turnkey price of the PV system is 30 000$, in addition of 5000$ to the cost of maintenance during the lifecycle of PV system (25 years). Japanese Grid connect Standard is used to connect PV system to Grid. Its ideal within the plus or minus 45 degrees from due south. Its ideal to set up the solar cell at due south. But, it can expect 96% amount of generation at the direction of south east and of southwest, if it can generate 100% at due south. For Hanoi area, the PV array is installed in southbound, tilted 50 (Fig.4) [4]
The Fig.5 shows the power production of PV system during one day. The power is high in the after-noon (from 10h to 15h) and it is proportional with solar irradiance. The total energy production for each month is presented in Fig.6. In summer, PV power production is higher than in the winter.
8000
7000
6000 5000 4000 3000 2000 1000 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
47
kWh 1,000.00
900.00
800.00 700.00 600.00 500.00 400.00 300.00 200.00 100.00
0.00
Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep 10 11 12 1 2 3 4 5 6 7 8 9 2012 2012 2012 2013 2013 2013 2013 2013 2013 2013 2013 2013
For instant, there is not Feed-in Tariff (FIT) for PV power generation in Vietnam. If the energy price of PV system is 0,075$/kWh (the energy price of EVN), the investors cannot recover capital (Fig.7.a).
If the energy price of PV system is 0.3$/kWh (energy price according to studies in National Master Plan on Renewable Energy Development of Vietnam), the investors need 14 years to recover capital (Fig.7.b).
Fig.8. Cumulative cash flows graph for the energy price 0.525 $/1kWh (installation in Japan)
48
Let us consider a similar PV system installed in Japan, we have the support measure by METI (Ministry of Economy, Trade and Industry): 800 $/1kW subsidy, Start of the feed-in Tariff (FIT) program in July 2012, 0.525$/1kWh for Grid-Connected PV System [2]-[3]. The Fig.8 shows that the investors need 7 years to recover capital and 20 years later they have 70 000$ profit
For the future, the hybrid power system will be used (Fig.9). Hybrid power systems are designed for the generation and use of electrical power. They are independent of a large, centralized electricity grid and incorporate more than one type of power source. They may range in size from relatively large island grids of many megawatts to individual household power supplies on the order of one kilowatt. In general a hybrid system might contain AC or DC diesel generators, an AC or DC distribution system, loads, renewable power sources (wind turbines, microturbines, or photovoltaic power sources), energy storage, power converters, rotary converters, coupled diesel systems, dump loads, load management options, or a supervisory control system.
4. CONCLUSION
In this paper, we have presented an overview of PV generator in Vietnam, Japan and in the world. This paper shows that the PV system cannot develop if there are not the suitable policies of the government. To ensure energy security, the PV development is necessary. Short-term, we need to have the incentive policies and the Vietnam Grid connect Standard based on the reference of the other country (Japan, France, and Germany).
REFERENCES
[1] INTERNATIONAL ENERGY AGENCY: A Snapshot of Global PV 1992-2012 [2] International Energy Agency-Photovoltaic Power Systems Program (IEA-PVPS), http://www.iea-pvps.org/ [3] Hiroyuki YAMADA, National Survey Report of PV Power Applications in Japan 2012, INTERNATIONAL ENERGY AGENCY CO-OPERATIVE PROGRAMME ON PHOTOVOLTAIC POWER SYSTEMS, 2012 [4] Collen OBrien, Roof-Mounted Solar Photovoltaic Arrays: Technical Issues for the Roofing Industry, RCI, Inc. 800 828-1902 www.rci-online.org
49
Session 1
Grid connected P' system facing %oltage sags I solutions to !%oid 2nwanted disconnection
#e Thi .inh Cha!( T!ng #e 8!c Hanoi University of Science and Technology, 0. *ai Co Viet, Hanoi, Vietnam
Abstract - !mong disturbances on &ower networ=3 grid-connected &$oto%oltaic ?P'@ in%erters are %ery sensiti%e wit$ %oltage sags t$at can cause disconnections of P' systems. "$is disconnection is sometimes unwanted3 in &articularly w$en t$e fault is situated on an ad7acent feeder su&&lied by t$e same substation. .uture &ower systems wit$ a large s$are of P' systems connected could be se%erely affected if se%eral of t$e P' systems are tri&&ing at t$e same instant. "$erefore3 t$e first ob7ecti%e in t$is &a&er is to identify unwanted disconnections of P' systems by analy:ing t$e be$a%ior of grid-connected P' systems facing %oltage sags caused by s$ort-circuits. "$en3 solutions to a%oid unwanted disconnections by using t$e %oltage-times c$aracteristics and modifying t$e time delay of t$eir decou&ling &rotections are &ro&osed. "$e %alidation by simulations s$ows t$e efficiency of t$e &ro&osed solutions. Index Terms - P' system3 'oltage sag3 !nti-islanding &rotection3 2nwanted disconnection.
50
AbstractAmong disturbances on power network, gridconnected photovoltaic (PV) inverters are very sensitive with voltage sags that can cause disconnections of PV systems. This disconnection is sometimes unwanted, in particularly when the fault is situated on an adjacent feeder supplied by the same substation. Future power systems with a large share of PV systems connected could be severely affected if several of the PV systems are tripping at the same instant. Therefore, the first objective in this paper is to identify unwanted disconnections of PV systems by analyzing the behavior of gridconnected PV systems facing voltage sags caused by shortcircuits. Then, solutions to avoid unwanted disconnections by using the voltage-times characteristics and modifying the time delay of their decoupling protections are proposed. The validation by simulations shows the efficiency of the proposed solutions. Index Terms PV system, Voltage sag, Anti-islanding protection, Unwanted disconnection.
LV networks). A low voltage caused by voltage sag can provoke a disconnection of PV systems due to their decoupling protection. The disconnection of significant amount of these PV systems could have local and global impact on the network operation, in particular on weak grids. This is why, it is necessary to find an efficient solution which avoids the unwanted disconnection of PV systems. The aim of this paper is to: - Identify the unwanted disconnections of PV systems - Propose solution for PV systems by taking into account the following requirements: PV systems must remain connected when a fault occurs on an adjacent feeder fed from the same substation but they must be disconnected in the case of a fault occurring in its zone (the fault on the feeder where the PV systems are connected). Firstly, this paper will present modeling of PV systems with decoupling protection, urban LV and MV networks and their protection systems. These models enable us to analyze behaviors of the PV system connected to distribution networks facing disturbances, especially the voltage sags caused by different types of short-circuit. Then, simulations will be carried out in order to identify the cases of unwanted disconnection of PV systems. Finally, solutions will be presented to avoid an unwanted disconnection of PV systems by using the voltage-times characteristics and modifying the time delay of their disconnection protections. II. A. MODELING
I.
INTRODUCTION
he connection of PV systems in the network requires the coordination of protections between the PV system with their disconnection protections and the networks protections. In order to ensure a good operation of PV systems and network, several requirements for the connection of PV systems have been issued. In France, in order to ensure a good operation of PV system protection systems, DIN VDE 0126 [1] requires that PV systems connected to LV network are disconnected within 0,2s if the voltage at the PV inverter terminal lower than 80% or greater than 115% of the nominal voltage. The decoupling protection for large PV plants connected to MV network is divided into different types called H.1, H.2, H.3, H.4, and H.5. The classification of these types is based on rated power of PV plant and on MV network configuration. Among disturbances on power network, voltage sags are known as one of the major causes of nuisance trips of PV systems. Recently, several works have been carried out to study the behaviors of PV systems facing voltage sags [2, 3]. A short-circuit can cause voltage sags, whose depth and duration vary and depend on network configuration, operating state and on eventual presence of Distributed Generation (DG). Faults can occur at any level of network (HV, MV or
Chau Le-Thi-Minh and Tung Le Duc is with the Hanoi University of Science and Technologie, 01 Dai Co Viet Hai Ba Trung Hanoi Vietnam.
1 1 + sT p
1 1 + sTq
Vmes
P Q
Pref Qref
Calculation
IDref IQref
PI
Calculation
PI
Dynamic 1 1 + sT
In this part, we present modeling of a three phase/single phase PV system. The PV system operates in classic mode assuring
51
that the reactive power (Q) is always assigned to 0. For the P/Q control scheme used for PV inverter (Figure 1), the active and reactive power outputs of the PV inverter are fixed to setpoint values Pref (in function of solar irradiation) and Qref (=0). The limit of current output (IDref and IQref) is set to 1.1 of its nominal value. B. Modeling distribution network protections In order to study the behaviors of PV inverters facing disturbances, protections of LV network, MV feeder and islanding protections of PV inverters are modelized. For LV network in France, the fault protection is assured by breakers and fuses. In our study, only the feeder fuse (FD 400A) and customers fuses (AD 90A) are used. The customer is protected by breaker and fuse AD and the LV network is protected by feeder fuse FD For MV network, an urban distribution network is used. There is not auto-recloser in each feeder. When the current exceeds the set point corresponding to these thresholds (Rephi > 0.8*Iscbi and Rh > 1.2*Icap), the protection activates and trips the feeder with the time delay of 500ms. The PV systems were also equipped with decoupling protection. Typically PV system models have both under/over voltage (OVP/UPV) and under/over frequency (OFP/UFP) protection. It means that the PV inverter will be disconnected if the frequency or amplitude of the voltage at the connecting point is outside the limits (0.85 and 1.15pu for voltage, 47.5 and 50.2 Hz for frequency). For further principles of and OVP/UPV protection methods, see [4]. In addition, zerosequence voltage protection (Vo>10%Vn) is used to detect the single phase fault. All PV systems connected to LV network respect the standard "DIN VDE 0126". So if the voltage caused by a short circuit is lower than 80% or greater than 115% of the nominal voltage, the decoupling protection of PV inverter is activated within 200ms. For the PV plant connected to urban MV network, the type of decoupling protection is H.1 and H.2 (the PV power is in the range of about 500kW to 2 MW) [5,6].
TABLE 1 TYPES OF COUPLING PROTECTION
urban LV network is connected at bus 24 from the MV network. This LV network contains a 20/0.4kV, 400kVA transformer, two feeders. The residential loads are modeled by single-phase loads (RL circuit). Single-phase PV productions are connected to buses 21, 23 and 24.
III.
Different types of short circuit and different fault positions are considered. For each fault position, the three following types of short-circuits are considered: - Three phase fault - Phase-to-phase, phase-to-phase to neutral and phase-tophase to ground fault - Phase to neutral and phase to ground fault. By using urban networks (MV, LV network) and the models (PV model, network protections, decoupling protections) developed in the preceding part, the behaviors of gridconnected PV inverters and the identification of unwanted disconnections of PV inverters are studied in several scenarios. The results obtained by simulations have shown that almost short-circuits occurring on the LV/MV adjacent feeder, disconnections of PV systems due to the voltage sags are unwanted. For example, a three-phase short circuit on the adjacent feeder L_05 (outside of PV zone) in MV network.
1 0.8 0.6 0.4 x 10
4
With: t0 is the maximum temoprisation of zero-sequene protection of MV feeder protection, about 1 or 1.9 seconds and t 1 is the maximum temoprisation of phase protection of MV feeder
C. Modeling the studied network An urban MV network is presented in figure 2. This network contains 6 feeders, supplied by a 63/20kV - 36 MVA transformer. Only the feeder L_06 was modeled in detail. The other feeders, used only for studying faults located in adjacent feeder, are modeled by a cable with an equivalent load. Two three phase PV systems are connected to buses 30 and 14. An
Curent(A)
0.2 0 -0.2 -0.4 -0.6 -0.8 -1 0.2 0.4 0.6 0.8 1 1.2
T ime ( s)
52
Figure 3 shows the current at the adjacent feeder L-05. The fault current increases very strongly and superior to the limit, and the adjacent feeders protection activates after 500ms.
500 400 300 200
100 0 -100 -200 -300 -400 0.2 0.4 0.6 0.8 1 1.2
Time(s)
Figure 4 presents the current at the feeder L_06 where PV systems are connected in LV network. This current is still within the limits, so L_06 feeders protection does not activate and this feeder maintain connected to the network.
0126. In this part, the solution is to integrate a voltage-time characteristic in the decoupling protection for all PV systems instead of the limits defined by DIN VDE 0126. In case of fault on the adjacent feeder L_05 in urban MV network, figure 6 shows the PV system voltage (PV connected to urban feeder L_06 in LV network) for a three-phase short circuit in comparison with the proposed voltage-time characteristic. During short circuit, the PV system voltage is greater than the voltage defined by the voltage-time characteristic. 500ms after occurrence of the fault, the fault in adjacent feeder is cleared by adjacent feeders protection (L_05). After this action, the voltage in feeder L_06 comes back to a normal value (L_06 feeders protection does not get activated) and feeder L_06 maintains connection to the network.
1.2 1 0.8 0.6 0.4 0.2 0 0.2
Temporisation du protection du dpart adjacent
1 0.5 0 0
Curent(A)
Voltage-time Characteristic Tension du PV Vpv 1 with using proposed solution Gabarit de tension 2 the 3
0.4
0.6
0.8
1.2
1.4
Figure 6. PV voltage in comparison with the voltage-time characteristic for a fault in MV adjacent feeder (feeder L_05, outside of PV zone)
Figure 5. Voltage of PV system in case of three phases fault
However, it causes a disconnection of all PV inverters connected in feeder L_06 by decoupling protection of PV inverters because the voltage at the PV connection bus is lower than 0.8pu (standard DIN VDE 0126). The figure 5 shows the PV voltage. For the PV system connected to MV network, similar results are obtained. When the voltage sag caused by short-circuit are below 85% nominal voltage, The PV system are connected instantaneous. The disconnection of PV inverters for faults in adjacent feeder in MV network is unwanted. Solutions must be found to avoid this disconnection. In the following part, we propose two solutions to ensure protection selectivity based on voltage-time characteristic and the temporisation of decoupling protection. IV.
SOLUTION TO AVOID UNWANTED DISCONNECTIONS FOR GRID-PV SYSTEMS
For urban networks, there is no auto-recloser in any feeder. When the current exceeds the protection relay thresholds, the feeder protection activates and trips the circuit breaker with the time delay of 500ms. Thus, for the PV connected to MV network, the solution is to increase time delay of decoupling protection integrated into the PV systems greater than 500ms, in this paper is t0 +0,5ms for detection of MV single phase fault and t1 +0,5ms for detection of MV multiphase fault. By increasing the protection time delay, the adjacent feeders protection has enough time to trip the fault before the disconnection protection of PV systems operates. The figure 7 shows the PV power in two cases with and without using the proposed solution.
In France, all distributed generations (DG) greater than 5MW and connected to the MV network must remain connected in case of voltage sags above a limit defined by a voltage-times characteristic [7]. This characteristic has a voltage threshold of 0.05 pu during 150 ms, then the voltage have a threshold of 0.2 pu during 350ms, and finally the voltage increases linearly to 0.9 pu during 1000 ms. The generator must remain connected when the voltage stays in the domain above the red curve in Figure 6. For the PV inverter connected to LV network, the existing disconnection protection operates with respect to DIN VDE
Figure 7. Power of 500kW PV system with type H.1 connected to feeder L_06 in MV network
In comparison with the results in the previous part, by applying these solutions, unwanted trips of PV systems are avoided. V. CONCLUSION
53
systems are very sensitive to voltage sags, and in some cases, there are unwanted disconnection of PV systems, in particular for a fault in LV and MV adjacent feeders. Therefore, solutions by using the voltage-time characteristic for PV systems connected to LV network and on modifying the temporisation of decoupling protection for PV systems connected to LV network are proposed. The validation by simulations shows the efficiency of the proposed solution which can avoid unwanted disconnection of PV systems in case of a short circuit outside of PV zones. VI.
[1] [2]
REFERENCES
Norme DIN VDE 0126-1-1, Automatic disconnection device between a generator and the public low-voltage grid, February 2006. B. Bletterie, R. Brndlinger, H. Fechner, Sensitivity of photovoltaic inverters to voltage sags Test results for a set of commercial products, 18th International Conference and Exhibition on Electricity Distribution, CIRED, Turin Italy, 2005. Botjan BLAIC*, Arsen JURASIC, Igor PAPIC, Simulating the dynamic response for a photovoltaic generation system to voltage sags, 18th International Conference and Exhibition on Electricity Distribution, CIRED, Turin Italy, 2005 Ward Bower and Michael Ropp, Evaluation of Islanding Detection Methods for Utility-Interactive Systems in Photovoltaic Systems, Report IEA PVPS T5-09: 2002 Technical reference, Study of the impact on the protection plan connecting a distributed generation in MV network, (ERDF-PRORES_09E) Technical reference, "Installation protection of the generation connected to the public distribution network" (ERDF-NOI-RES_13E) Arrt du 23.04.2008 relatif aux prescriptions techniques de conception et de fonctionnement pour le raccordement un rseau public de distribution dlectricit en basse tension ou en moyenne tension dune installation de production dnergie lectrique
[3]
[4]
[5]
[6] [7]
54
Session 1
3