Advanced Simulation tools in support of the design of the Power System architecture of the European Spallation Source (ESS) 24.10.2013 Prof.Monti, Junjie Tang E.ON Energy Research Center, RWTH Aachen University, Aachen, Germany The E.ON Energy Research Center June 2006: the largest research co-operation in Europe between a private company and a university was signed Five new professorships in the field of energy technology were defined across 4 faculties Research areas: energy savings, efficiency and sustainable power sources 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | E.ON ERC Infrastructure 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Outlines Background Simulation Project Goal Simulation Approach and Modeling Test Scenarios and Results Conclusions and Future works 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Background ESS: European Spallation Source ≡ A joint European project, which has partners from 17 European countries ≡ Trend for reduction of energy consumption and greenhouse gas (GHG) emission ≡ Design proposal: establish an energy concept for demanding energy targets to be = RESPONSIBLE – 20 per cent reduction in energy consumption = RENEWABLE – 100 per cent utilisation of renewable energy = RECYCLABLE – 60 per cent recycling of utilized energy Grid simulation project ≡ Sustains for 2 years ≡ Partners include: 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Background Fig.2: Process of generating neutron Fig.1: Structure of ESS and operation [source:http://europeanspallationsource.se/photos-images] Internal ESS distribution grid, connected with several sections Linear accelerator (LINAC), the key component in ESS facility 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Simulation Project Goal Develop a simulation model for the regional electricity grid, including ESS grid Evaluate the mutual impact between newly integrated ESS and the regional grid: ≡ predict disturbances that ESS can cause to the regional grid ≡ predict disturbances in the operation of the ESS that can be caused by the regional electricity grid Lund Grid 145 kV D G D G D G T1-ESS Wind farms Other loads in Lund 23 kV G T2-ESS ESS energy solution Evolution 24.10.2013 | ACS Automation of Complex Power Systems | G Regional power plant ESS grid Conventional energy solution Junjie Tang D G The RT Simulation Lab 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Co-simulation Approach—Introduction Real time simulation and Hardware In the Loop methods are essential tools for the development of future complex power systems Given the complexity of such a scenario the use of a single tool is unfeasible Co-simulation approaches have to be developed, so that: ≡ ≡ ≡ ≡ Dedicated tools and library can be shared Different expertees can be capitalized Facilities at different geographical locations can be interconnected Multi-rate execution can be perfomed 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Co-simulation Approach RTDS has a good ability to simulate electromagnetic transient, while Opal RT is compatible with the models built in MATLAB/Simulink The idea is to remain Lund grid models in RTDS, while to model ESS grid in Opal RT including the power electronics of LINAC Connection between two simulators for co-simulation by analog interface 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Modeling Lund Grid in RTDS Rack 1 Current source controlled by Opal RT signal Rack 2 Fig.: Lund grid modeling in RTDS 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Modeling ESS internal grid in MATLAB/Simulink Modeling Power electrical part Transplant Download I/O ports Exchange with RTDS Power eletronics part 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Modeling ESS internal grid in RT-LAB Subsystem 2 of ESS grid Subsystem 3 of ESS grid Subsystem 1 -equivalent voltage source and LINAC loads Voltage source controlled by RTDS signal Subsystem 4 of ESS grid 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | ESS grid is split to 4 subsystems Each subsystem using one CPU/core for realtime simulation with parallel technology Modeling power electronics for LINAC in RT-LAB Power electronics part amplify to 200 times P: variation in range of [75, 310] kW P: variation in range of [15, 62] MW Depends on control of power electronics, the power consumption is not very close to constant Simulate 200 LINACs with power electronics in real-time is infeasible Equivalent modeling with a controllable load and a amplified signal, is used to simulate power amplification 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Testing Scenarios—Short circuit Test scenarios Location Type Conditions, purpose ESS_in three phase L-G with different fault clearing times: 0.05 s, 0.1 s and 0.3 s ESS_in three phase L-G with different impedances of T1-ESS and T2-ESS: 11.5%, 15% and 20% ESS_in three phase L-G with different short circuit impedance: 0.01 Ω, 0.1 Ω and 1Ω ESS_in Phase A and B L-L compare with three phase L-G ESS_in Phase A L-G compare with three phase L-G ESS_out three phase L-G with different impedances of T1-ESS and T2-ESS: 11.5%, 15% and 20% ESS_out three phase L-G with different fault clearing times: 0.05 s, 0.1 s and 0.3 s ESS_out Phase A and B L-L compare with three phase L-G ESS_out Phase A L-G compare with three phase L-G SEE three phase L-G compare faults at different buses SEE Phase A and B L-L compare with three phase L-G at SEE SEE Phase A L-G compare with three phase L-G at SEE VKA three phase L-G compare faults at different buses Short circuit To check the impacts when various types of short circuits happen at different locations, as well as impedances of transformers and grounding 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Testing Scenarios—Loss of components Test scenarios Tripping of power source Loss of branch and transformer Location Type HVA HVB HVDC Wind farm Örtofta and Helene simultaneously ÖVT Steam and Gas simultaneously (SEE-ESS) and (ESS-Lund) simultaneously (SEE-ESS) and (HVA-BBK) simultaneously (SEE-ESS) and (SEE-VKA) simultaneously T2-ESS tripping tripping tripping tripping tripping Purpose compare such kind of faults at different buses tripping loss loss compare such kind of faults at different buses loss loss As common fault types - loss of components With cascading faults occurring more frequently than in the past, different combinations of component loss are relevant Loss of transformers connecting ESS and Lund grid - possible disruptive impact on the operation of ESS 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Testing Scenarios—Normal operations Test scenarios Location Connection ESS_in Disconnection ESS_in Charging schemes ESS_in compare with the impact caused by charging scheme I and III on grid ESS_in analysis for realistic data based harmonics Harmonics check the impact of such operations on the grid ESS_in SEE Maxlab Wind farm Purpose Lillgrund analysis for the threshold of THD 8% (as defined in the grid code of E.ON Sverige AB) based harmonics check the impact of wind farm with different generation ratio: 0%, 50%, 100%, in aspect of power flow Normal operation of ESS may also impact the whole grid Charging schemes of LINAC mainly determine the load characteristics of ESS Variability of wind power generation impacts power flow and state variation in Lund grid 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Test Result 1 3-phase L-G short circuit at ESS_in with different transformers impedances 0 0.1 0.2 0.3 0.4 0.5 t [s] 0.6 0.7 0.8 Phase A Phase B Phase C 0.9 1 50 0 -50 0 0.1 0.2 0.3 0.4 0.5 t [s] 0.6 0.7 50 0 0.2 0.3 0.4 0.5 t [s] 0.6 0.7 ESS__out ESS__in Maxlab SEE 50.5 50 49.5 0 0.1 0.2 0.3 0.4 0.5 t [s] 0.6 0.7 50 0 0.1 0.2 0.3 0.4 0.5 t [s] 0.6 0.7 0.8 0.9 50 0 0.1 0.2 0.3 0.4 0.5 t [s] 0.6 0.7 0.8 0.9 0 1 1 0.1 0.2 0.3 0.4 0.5 t [s] 0.6 0.7 0 0.8 0.9 1 ESS__out ESS__in Maxlab SEE 0 0.1 0.2 0.3 0.4 0.5 t [s] 0.6 0.7 0.8 0.9 1 ESS__out ESS__in Maxlab SEE 100 1 ESS__out ESS__in Maxlab SEE 51 49 0.9 ESS__out ESS__in Maxlab SEE 50.5 49.5 0.8 0 100 Angle [radians] 0.1 200 0 Angle [radians] 0 200 ESS__out ESS__in Maxlab SEE 100 Angle [radians] -50 |Voltage| [kV] 0.8 Phase A Phase B Phase C 0.9 1 0 -50 200 |Voltage| [kV] 0.8 Phase A Phase B Phase C 0.9 1 |Voltage| [kV] Voltage of buses 50 Frequency of buses Frequency [Hz] Frequency [Hz] Frequency [Hz] Iabc [kA] Iabc [kA] Iabc [kA] Short circuit current with transformer impedance 11.5% , 15% , 20% 0 0.1 0.2 0.3 0.4 0.5 t [s] 0.6 0.7 0.8 0.9 1 0.7 ESS__out ESS__in Maxlab SEE 0.8 0.9 1 0.7 ESS__out ESS__in Maxlab SEE 0.8 0.9 1 Angle of buses 2 0 -2 0 0.1 0.2 0.3 0.4 0.5 t [s] 0.6 2 0 -2 0 0.1 0.2 0.3 0.4 0.5 t [s] 0.6 ESS__out ESS__in Maxlab SEE 2 0 -2 0 0.1 0.2 0.3 0.4 0.5 t [s] 0.6 0.7 0.8 0.9 1 With different impedances of transformers 11.5%, 15% and 20% Maximum of AC short circuit current are 30.16, 26.54 and 21.59 kA individually Higher impedance of the transformers can restraint the short circuit if such kind fault happens at ESS internal 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Test Result 2 Loss of transformer T2-ESS Voltage when loss of T2-ESS Frequency when loss of T2-ESS 142 50.05 140 50 138 ESS__out ESS__in Maxlab SEE 134 49.95 Frequency [Hz] |Voltage| [kV] 136 132 130 49.9 49.85 128 ESS__out ESS__in Maxlab SEE 126 49.8 124 122 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 t [s] Three phase voltage of ESS-in 49.75 1 0 0.5 0.4 0.3 0.2 0.1 t [s] Three phase voltage of ESS-out 0.6 0.7 0.8 0.9 1 Phase A Phase B Phase C 20 0 -20 100 Voltage [kV] Voltage [kV] 40 Phase A Phase B Phase C 50 0 -50 -100 -40 0.05 0.1 0.15 0.2 t [s] Three phase voltage of Maxlab 0.25 0.05 0.1 0.15 0.2 t [s] Three phase voltage of SEE 0.25 Phase A Phase B Phase C 10 0 -10 -20 0.05 0.1 0.15 t [s] 0.2 100 Angle [radians] Voltage [kV] 20 0.25 Phase A Phase B Phase C 50 0 -50 -100 0.05 0.1 0.15 t [s] 0.2 0.25 Maximum power consumption of ESS 47 MW, while capacity of transformer 40 MVA Once any one of the transformers is out of work, there is a potential risk of overload for the other one For the voltage of ESS_in, the decrease is over 10% 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Test Result 3 Short circuit at ESS_in with different fault types Three phase voltage of ESS-in Three phase voltage of ESS-out Three phase voltage of ESS-in -20 Phase A Phase B Phase C 50 0 -50 -100 0.1 0.15 0.2 t [s] Three phase voltage of Maxlab 0.25 0.05 Phase A Phase B Phase C 20 0 -20 100 Voltage [kV] 0 100 Voltage [kV] 20 -40 0.05 Phase A Phase B Phase C 50 0 -50 -100 0.1 0.15 0.2 t [s] Three phase voltage of SEE -40 0.05 0.25 20 0.1 0.15 0.2 t [s] Three phase voltage of Maxlab 0.25 0.05 0.1 0.15 0.2 t [s] Three phase voltage of SEE 0.25 -10 Phase A Phase B Phase C 50 0 -50 Phase A Phase B Phase C 10 0 -10 -100 -20 0.05 0.1 0.15 t [s] 0.2 0.25 0.05 0.1 Single phase A L-G Three phase voltage of ESS-in 0.15 t [s] 0.2 -20 0.05 0.25 0.1 0.15 t [s] 0.2 0.25 100 Angle [radians] 0 100 Voltage [kV] 20 Phase A Phase B Phase C 10 Angle [radians] Voltage [kV] Three phase voltage of ESS-out 40 Phase A Phase B Phase C Voltage [kV] Voltage [kV] 40 Phase A Phase B Phase C 50 0 -50 -100 0.05 0.1 Phase AB L-L 0.15 t [s] 0.2 0.25 Three phase voltage of ESS-out Phase A Phase B Phase C 20 0 -20 100 Voltage [kV] Voltage [kV] 40 Phase A Phase B Phase C 50 0 -50 -100 -40 0.05 0.1 0.15 0.2 t [s] Three phase voltage of Maxlab 0.25 0.05 0.1 0.15 0.2 t [s] Three phase voltage of SEE Fault type Va (kV) Vb (kV) Vc (kV) Single phase A L-G 0.3778 28.04 31.48 Phase AB L-L 8.94 7.48 16.37 Three phase L-G 2.14 2.06 2.39 0.25 Phase A Phase B Phase C 10 0 -10 100 Angle [radians] Voltage [kV] 20 Phase A Phase B Phase C 50 0 -50 -100 -20 0.05 0.1 0.15 t [s] 0.2 0.25 0.05 0.1 Three phase L-G 0.15 t [s] 0.2 0.25 Recordings about the instantaneous voltage at ESS_in, ESS_out, Maxlab and SEE Single phase L-G short circuit introduces overvoltage to the other two phases of ESS_in Two-phase L-L short circuit lead voltage decrease in the two phases of ESS_in Three-phase L-G cause serious undervoltage for each phase of ESS_in 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Test Result 4 3-phase L-G short circuit at different locations in the Lund grid Bus_Name AIE BBK BFO ELVS ESS HVA HVB HVDC Helene Lund ÖM MRP Maxlab ÖVT Örtofta SEE SÖV VKA VPE HRGD Hörby Lillgrund SSY Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Fig.: Maximum instantaneous amplitudes of three-phase L-G short circuit current of each bus when ESS is connected to Lund grid Three-phase L-G short circuit is the most detrimental to the system Effects of short circuit faults highly depend on their locations in the grid The fact that the ESS is connected or disconnected from the Lund grid has a minor impact on the short circuit current 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Test Result 5 Impact from different charging schemes of LINAC of ESS grid Active Power Consumption [MW] 55 Charging scheme I Charging scheme III 50 45 40 35 30 25 20 15 0 0.2 0.4 0.6 0.8 1 t [s] 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Test Result 5 Impact from different charging schemes of LINAC of ESS grid Scheme I: |Voltage| [kV] 600 ESS__out ESS__in Bus 12 Bus 15 400 200 0 0 0.1 0.2 0.3 0.4 0.5 t [s] 0.6 0.7 0.8 0.9 1 Scheme II: |Voltage| [kV] 145 140 ESS__out ESS__in Bus 12 Bus 15 135 130 125 0 0.1 0.2 0.3 0.4 0.5 t [s] 0.6 0.7 0.8 0.9 1 Voltage: the Charging Scheme I yields high flicker to the other loads, especially for Bus 12 (another research facility around) Frequency of the voltage pulses is identical to charging frequency of Scheme I (constant current) With respect to the voltage, Charging Scheme I is worse than Charging Scheme II (constant power) 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Test Result 6 Wind farm with different generation ratios from 0% to 50% from 50% to 100% Due to the power being mostly provided by transmission network, the impact aroused by the wind farms is limited, from the point view of power flow distribution 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Test Result 7 Uncertainty analysis for the variation of loads and wind farm generation Load condition Maximum Minimum Substation (MW) (MW) AIE 400 45 BBK 75 25 BFO 15 4 ELVS 55 10 Heleneholm 25 10 Lund ÖM 35 15 Maxlab 20 15 MRP 110 25 SEE 235 100 SÖV 15 2 VKA 15 4 VPE 110 50 ÖVT 10 2.5 ESS (Charging 45.7 31.7 Scheme II) Test scenario Uncertainty source 1 Load variation of Bus AIE, MRP, SEE and VPE 2 Intermittence of the two wind farms 3 Load variation of ESS 4 Load variation of the four buses in test scenario 1 together with ESS, and intermittence of the two wind farms Loads vary with uniform distributions, while the Weibull distribution for wind speed of wind farms Uncertainty sources include five selected load buses with large loads and two wind farms RTDS simulation co-operated with Monte Carlo (MC) method is adopted to investigate the uncertainty issue, (10000 MC simulations) 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Test Result 7 Uncertainty analysis for the variation of loads and wind farm generation Scenario 1 Scenario 2 Scenario 3 Scenario 4 Ranges of voltage variation in scenarios are different Some voltages even drop to an unacceptable range in test scenario 2 and 4 due to the integration of wind farms Voltage is more volatile at the buses located with wind farm or heavy loads 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Test Result 7 Uncertainty analysis for the variation of loads and wind farm generation 0.06 0.4 0.04 0.02 0 137 0.2 Bus 15 Density Density scenario 1 scenario 2 scenario 4 0.3 0.15 0.1 0.05 137.5 138 Voltage [kV] 0 140.5 138.5 0.4 141 141.5 Voltage [kV] 0 136 0.3 0.2 0.1 137 138 139 140 Voltage [kV] 141 142 Fig.1: Frequency histogram of voltage at Bus 5 in test scenario 1, 2, 4 respectively. 0 141 Bus 21 Density 0.1 142 0.2 Bus 14 Density Density 0.2 Bus 12 0.15 0.1 0.05 141.5 142 Voltage [kV] 142.5 0 140.5 141 141.5 Voltage [kV] Fig.2: Frequency histogram of voltage at Bus 12, 14, 15 and 21 in test scenario 3 Uncertainty merely from wind farm generations does not yield a large variation to the voltage Bus 5 (ESS), while variations from the four largest load buses lead to a broader range of possible voltage values Uncertainty of wind farms can mitigate the impact from stochastic load variations In terms of voltage variation, Bus 12 (Maxlab) is larger in comparison with Bus 14 (power plant), Bus 15 (big load consumer) and Bus 21 (wind farm) 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang 142 | Test Result 8 Uncertainty analysis for the variation of loads in future scenarios 2013 Inauguration Now 300 250 250 250 150 100 50 Frequency histogram [p.u.] 300 200 200 150 100 50 0 136 137 138 139 Voltage of ESS [kV] 140 141 0 136 Time Fully in operation 300 Frequency histogram [p.u.] Frequency histogram [p.u.] 2025 2019 200 150 100 50 137 138 139 Voltage of ESS [kV] 140 141 0 136 137 138 139 Voltage of ESS [kV] 140 Load demand is uniform distributed, and its growth rate is assumed 5% per year Errors in load prediction following a normal distribution (0, 0.01), the incrementals are 0, 30%*(1±0.03) and 60%*(1±0.03) individually Uncertainty increases along with time, the impact will be bigger and more chanllenging in future scenarios 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | 141 Conclusions Main transformers connecting ESS grid and Lund grid highly decide the status of ESS grid Load characteristics of ESS grid mainly depends on the charging schemes of LINAC, which is possible to disturb the power quality of Lund grid Integration of ESS brings a minor impact on short circuit current of the buses in Lund grid Involvement of wind farms has a slight influence on the operation of ESS grid and Lund grid A deeper awareness about how uncertainty from the renewable sources and loads affects the operation of ESS grid and Lund grid is obtained 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Future works Heat energy can be also considered in ESS grid, as well as Lund district heating system, will be modeled in RT-LAB to evaluate the energy efficiency from a global view in co-simulation A platform is under construction for combining real-time simulation and uncertainty quantification, to provide a possibility to extend such analysis even to Hardware in the Loop and Power Hardware in the Loop tests 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang | Thanks for your attention! Any question? 24.10.2013 | ACS Automation of Complex Power Systems Junjie Tang |