ELECTRA IRP Use Cases Simulations

Transcription

1 Designing and validating the future, intelligent, electric power systems Kassel, Sept. 6 th 2017 ELECTRA IRP Use Cases Simulations Julia Merino Fernández The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/ ) under grant agreement n

2 Overview PPVC Adaptive FCC BRC+FCC = DLFC IRPC vs. Fast FCC Fundamentals Setup Simulation results 2

3 PPVC 3

4 PPVC Fundamentals Two operation modes Proactive Corrective Planning phase: Proactive Real-time operation: Corrective Proactive 4

5 A PPVC single-cell test setup European CIGRÉ MV grid benchmark (modified with DERs) Load, PV pannels DERs patterns adjusted to match the grid requirements and dynamic controls have been modeled PV panels generation forecast Two feeders DERs PQ load forecast 8 PV plants 2 batteries 2 fuel cells 2 CHPs 1 Wind Turbine Note: Implementation in PowerFactory software 5

6 A) Planning phase ( Proactive mode) PPVC Results (I) Optimal voltage set-point are calculated for the next 15-min windows in 1-min intervals Why optimal power flows in a Planning phase? The advantages of the WoC Voltage control optimization in a single step Full observability Advanced communication technologies (less communication efforts) Increase in the calculation speed and data storage managment capabilities Total P losses = MW Total Q losses = MVAr Total P losses = MW (-4,3574%) Total Q losses = MVAr ( %) Non-optimized planned voltage in t є (t A+1 t A+2 ) (current practice) Optimized planned voltage in t є (t A+1 t A+2 ) (WoC proposed practice) 6

7 PPVC results (II) B) Real-time operation ( Corrective mode) Real node voltages An unexpected event occurs during the real-time cell operation (t A ) and the previous calculated pattern (in the proactive mode) needs to be recalculated Optimal set-points before/after the event En t=t A, U _NODE12 <lower limit_safe band 7

8 B) Proactive vs. Corrective mode PPVC results (III) 8

9 The WoC GS PV WP 9

10 Adaptative FCC 10

11 Adaptive FCC fundamentals Frequency Droop responsibility explicitly delegated to cells Optimal decomposition of the cell s CPFC (contribution to NPFC) over the available resources adding a droop scaling factor that depends on the state of the cells Avoid that FCC activations cause imbalances Responsibilization 11

12 A-FCC test setup Test assumptions CIGRE MV reference grid 1HV and 3MV cells Two adaptive controllers in cells 1 and 2 Only the HV cell was equipped with frequency restoration process Test Grid setup Rule table ΔP tie,i Δf i NH NL ZE PL PH NH 0% 33% 100% 100% 100% NL 33% 66% 100% 100% 100% ZE 66% 100% 100% 100% 66% PL 100% 100% 100% 66% 33% PH 100% 100% 100% 33% 0% 12

13 A-FCC test results Short-term performance (200s) 3MW increase in Node 02 Long-term performance (24-hour tests) Small droop of 1MW/Hz in the HV cell / Load step reductions every 15 min Overall use of reserves Function With Adaptive FCC With Fixed Droop ABS 8.59e+7(Ws) 1.07e+8(Ws) (Reduction) -19.7% Cost 6.94e+12(W 2 s) 9.46e+12(W 2 s) (Reduction) -26.7% Reduction is use of renewables RES With Adaptive FCC With Fixed Droop WG 1.24e+7(Ws) 1.51e+7(Ws) (Reduction) -18.0% PVs 8.97e+6(Ws) 9.56e+6(Ws) (Reduction) -6.2% ABS = P droop,1 + + P droop,n dt 0 Cost = P droop, Pdroop,n 2 dt 0 Max. Freq F-FCC = Hz M. Freq. A-FCC = Hz Min. Freq F-FCC = Hz M. Freq. A-FCC = Hz 13

14 FCC+BRC = DLFC 14

15 DTLC Fundamentals Currently employed AGCs use PI controllers Rigid tuning, susceptible to parameter variations Non-linearities (ramp rates, delays) problematic, only working points can be optimally tuned Proposed LFC Supported by the Web-of-Cells (WoC) capabilities: high degree of observability and comm. Two-staged control Power matching through direct observations Primary reference frequency control to balance unobserved powers and inaccurate measurements Tuning-free, adaptive, handles variable droops Stable over a wide range of parameters, agnostic to actuator non-linearities WoC s keep local problems local Secondary response decoupled from the system response Primary resources actively involved in frequency control Frequency is treated as a local quantity Inferred over primary resources states 15

16 AGC vs. DLFC 16

17 DTLC simulation results Scenario Parameter Cell 1 Cell 2 Cell 3 High J J [kg m²] H [s] P_pri_ramp [kw/s] T_sec_delay [s] Low J J[kg m²] H [s] P_pri_ramp[kW/s] T_sec_delay [s] Controller Parameter Cell 1 Cell 2 Cell 3 AGC T_ctrl [s] KP [1] KI [1/s] BRC T_ctrl [s] omega_f 1/3 1/3 1/3 omega_p 1/3 1/3 1/3 All load events (observed and unobserved) are 5 kw 17

18 DTLC conclusions The DLFC shows satisfying performance AGC only performing better in working points it was tuned against Non-linearities are handled very well, stability was proven on paper and in real experiments Frequency control works also under partial observability DLFC fits well into the BRC/WoC context Supports adaptive droop (FCC) Benefits from fast secondary response (BRC variant) Known shortcomings Good knowledge of controlled devices droop capability needed (but AGC needs it for the whole area) Can only balance tielines in case of total observability Tieline imbalance observer needed 18

19 IRPC+FCC Overview IRPC vs. Fast FCC 19

20 IRPC vs. FCC Overview This study aims at evaluating the capability of frequency containment control (FCC) and Inertia response control (IRPC) in mitigating the RoCoF as well as the two controllers effects on the frequency performance. The study is divided into two parts: Simulations study in DIgSILENT PowerFactory Experimental validation in SYSLAB (PowerLab DK) 20

21 IRPC vs. FCC setup 21

22 IRPC vs. FCC Controllers 2% droop Measurement delay (100ms) Device delay (10ms, 50ms, 100ms, 250ms) RoCoF calculation and Measurement delay (100ms) Device delay (10ms, 50ms, 100ms, 250ms) 22

23 IRPC vs. FCC results EVs are used as flexibility resources 2 kw load event IRPC Applied droops Frequency standard deviation for the two controllers Different granularity from EVs (i.e. EVs charging current are modulated with 1A granularity) Variation of Load steps 23

24 Can Fast frequency containment control compensate or replace the need for Inertia response control? This work analytically showed the interdependence between frequency containment and synthetic inertia control on the transient frequency variation and the RoCoF On the simulation level, it presented the ability of fast frequency control in improving the frequency in terms of nadir, steady state value and RoCoF. It also presented the ability of synthetic inertia control (IRPC) to improve the frequency nadir and slope following an event. Finally, an experimental validation was conducted, presenting the capabilities and limitations of the two controllers under two different circumstances: following load events in both directions and exogenous wind generation profiles 24

25 CONTACT INFORMATION Julia Merino Fernández, Ph.D. ELECTRA IRP website Link: Event details date and Place 25