2014 PV Distribution System Modeling Workshop: Determining Recommended Settings for Smart Inverters: Jeff Smith, EPRI
2 likes1,513 views
The document discusses the evaluation of smart inverter settings to enhance feeder voltage performance in relation to varying loads and solar PV conditions. It presents an analysis of different feeder sites, each with unique characteristics, and employs simulations to determine optimal smart inverter functions for specific objectives. Key insights include the challenges in identifying the 'best' settings and the impact of operational parameters on voltage stability and efficiency.
2014 PV Distribution System Modeling Workshop: Determining Recommended Settings for Smart Inverters: Jeff Smith, EPRI
- 1. Jeff Smith, Matt Rylander, Huijuan Li EPRI EPRI Smart Inverter Workshop, Santa Clara, CA 5/7/2014 Determining Recommended Settings for Smart Inverters
- 2. 2© 2014 Electric Power Research Institute, Inc. All rights reserved. Overview Objective Determine recommended settings for field site demonstration Evaluate the effectiveness of various smart inverter functions and settings for improving feeder voltage performance as load and PV vary over time Approach Time-series simulations in OpenDSS comparing feeder performance with and without smart inverter functions Sites Three different feeders, each with unique characteristics and overall objectives for use of smart inverters
- 3. 3© 2014 Electric Power Research Institute, Inc. All rights reserved. Which Smart Inverter Setting is Most Appropriate for My Situation? 0 5 10 15 20 25 1.024 1.026 1.028 1.03 1.032 1.034 1.036 1.038 1.04 1.042 1.044 Hour Voltage(pu) Voltages with different voltvar settings ---- Voltvar ---- No PV ---- PV base 115 unique volt/var control settings
- 4. 4© 2014 Electric Power Research Institute, Inc. All rights reserved. Site Characteristics Site kWdc (Panel size) kWac (inverter rating Short- circuit MVA @ POI X/R @ POI J1 1900 1700 30-36* 1.8-2.6 E1 605 566 38 1.8 H1 1000 1000 71 1.7 *multiple POI
- 5. 5© 2014 Electric Power Research Institute, Inc. All rights reserved. Overall Approach • Solar variability conditions – Clear day – Overcast day – Highly variable day • Load variability conditions – Peak load day – Minimum load day • Smart inverter settings – Volt-var – Volt-watt – Off-unity power factor 0 2 4 6 8 10 12 1 3 5 7 9 11 13 15 17 19 21 23 25 Power (MW) Local Time (Hour) Offpeak Peak Sandia’s variability index (VI) and clearness index (CI) to classify days Consideration for Different Feeder Load Profiles
- 6. 6© 2014 Electric Power Research Institute, Inc. All rights reserved. Smart Inverter Settings Power Factor Settings (inductive) 0.99 0.98 0.97 0.96 0.95 0.94 0.93 0.92 0.91 0.9 Sample volt/var curves shown: see Video for complete set of curves Similar range of curves used for volt/watt control
- 7. 7© 2014 Electric Power Research Institute, Inc. All rights reserved. Feeder Model Validation 0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 0 100 200 300 400 500 600 1 14 27 40 53 66 79 92 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 365 378 391 404 417 430 443 456 469 482 495 per‐unit voltage P (kW) P (kW) V_model (pu) V_measure (pu) E1 J1 H1 0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 0 200 400 600 800 1000 1200 1 16 31 46 61 76 91 106 121 136 151 166 181 196 211 226 241 256 271 286 301 316 331 346 361 376 391 406 421 436 451 466 481 496 per‐unit voltage P (kW) P (kW) V_measure (pu) V_model(pu) 0 50 100 150 200 250 1.02 1.03 1.04 1.05 1.06 1.07 0 50 100 150 200 250 300 PV (kW) Voltage (Vpu) Time (sec) Measured Simulated PV (kW)
- 8. 8© 2014 Electric Power Research Institute, Inc. All rights reserved. Smart Inverter Model Validation OpenDSS Simulations 260 265 270 275 280 285 290 295 Voltage (Vln) Time (s) Measured Simulated ‐400 ‐300 ‐200 ‐100 0 100 200 300 400 Reactive Power (kvar) Time (s) Measured Simulated
- 9. 9© 2014 Electric Power Research Institute, Inc. All rights reserved. Selecting the “Best” Smart Inverter Settings • Objectives – Each feeder analysis has unique set of objectives – Voltage – Efficiency – Control • Metrics – Approximately 20 conditions are monitored for each feeder – Only daylight impact is analyzed – Mean voltage at the point of common coupling (PCC) – Voltage variability index at the PCC – Tap operations – Losses • Rank objective impact based on the metrics for each scenario – Solar – Load 6 combinations all weighted equally (for now…)
- 10. 10© 2014 Electric Power Research Institute, Inc. All rights reserved. Demo Site J1: Objectives & Metrics Objective Metric Weight 1. Avoid overvoltage conditions 100 2. Improve customer efficiency 100 3. Reduce line regulator tap changes 100 4. Combined 1, 2, and 3 33/33/33
- 11. 11© 2014 Electric Power Research Institute, Inc. All rights reserved. Sample Plots Clear Day Overcast day Highly variable day PCCvoltage hour PCCvoltage hour PCCvoltage hour 0 5 10 15 20 25 30 1.02 1.025 1.03 1.035 1.04 1.045 1.05 0 5 10 15 20 25 30 1.02 1.025 1.03 1.035 1.04 1.045 1.05 0 5 10 15 20 25 30 1.01 1.02 1.03 1.04 1.05 1.06 1.07
- 12. 12© 2014 Electric Power Research Institute, Inc. All rights reserved. Circuit Performance Characterization
- 13. 13© 2014 Electric Power Research Institute, Inc. All rights reserved. Volt/Var Results Demo Site J1 Lesson Learned “best” settings can be difficult to identify 1.01 1.015 1.02 1.025 1.03 1.035 1.04 1.045 1.05 1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 Feeder Head Voltage Max Feeder Head Voltage (pu) Min Feeder Head Voltage (pu) 0 100 200 300 400 500 600 700 800 900 1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 103 109 115 Reg/LTC Tap Operations Tap Operations 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106 113 Cap Operations Cap Operations 1400 1450 1500 1550 1600 1650 1700 1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 Feeder Losses (kWh) Feeder Losses (kWh) 0.99 1 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 PCC Voltage Max PCC Voltage (pu) Min PCC Voltage (pu) 0 2 4 6 8 10 12 14 16 18 1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 103 109 115 VI at PCC VI at PCC 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106 113 Feeder End Voltage Max Feeder End Voltage (pu) Min Feeder End Voltage (pu) 0 1000 2000 3000 4000 5000 6000 7000 1 9 17 25 33 41 49 57 65 73 81 89 97 105 113Time Above ANSI (sec) Time Above ANSI (sec) 0.9 0.92 0.94 0.96 0.98 1 1.02 1.04 1.06 1.08 1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 103 109 Overall Feeder Min/Max Voltage Max Feeder Voltage (pu) Min Feeder Voltage (pu) Peak load day
- 14. 14© 2014 Electric Power Research Institute, Inc. All rights reserved. Demo Site J1 Combined Objective 4 Best Settings • Objective – Avoids overvoltage – Improves efficiency – Reduces tap operations • Metrics – Lower mean voltage – Flatter voltage profile – Less tap operations General trends in rank are due to rolling through different setting characteristics Lesson Learned Overall best settings have similar curves
- 15. 15© 2014 Electric Power Research Institute, Inc. All rights reserved. Sample Results – Volt/var control Demo Site J1 ‐1.2 ‐1 ‐0.8 ‐0.6 ‐0.4 ‐0.2 0 0.95 0.97 0.99 1.01 1.03 1.05 % Avail vars per‐unit voltage 1.01 1.02 1.03 1.04 1.05 1.06 1.07 0 5 10 15 20 25 30 per‐unit voltage Hour no_PV PV base voltvar ‐100 0 100 200 300 400 500 600 700 800 900 0 5 10 15 20 25 30 tap operations hour Tap_noPV Tap_Pvbase Tap_voltvar ‐1.5 ‐1 ‐0.5 0 0.5 1 1.5 0.95 0.97 0.99 1.01 1.03 1.05 1.07 1.09 Negative impact on voltage and line regulator operations Positive impact on voltage and line regulator operations 1.01 1.02 1.03 1.04 1.05 1.06 1.07 0 5 10 15 20 25 30 per‐unit voltage hour no_PV PV_base Voltvar Volt/var curve Daily voltage profile Regulator tap operations Volt/var curve Daily voltage profile Regulator tap operations 0 100 200 300 400 500 600 700 800 0 5 10 15 20 25 30 tap operations Tap_noPV Tap_Pvbase Tap_voltvar Lesson Learned Slight variation in settings can yield significantly different responses
- 16. 16© 2014 Electric Power Research Institute, Inc. All rights reserved. Demo Site J1 Trends in Volt/var Characteristics Best curves begin absorbing reactive power at 1.02 Vpu Best curves have a steep volt-var slope Lesson Learned Initial results indicate trends can be seen
- 17. 17© 2014 Electric Power Research Institute, Inc. All rights reserved. Demo Site J1 Best Setting Impact for Objective 4 • Each smart inverter function has one “Best” setting • Totalized metric for each “Best” setting in Objective 4 is shown • The Volt-var function and setting has the best impact for each metric PV volt/var volt/watt power factor PCC Mean Voltage (pu) 1.031 1.027 1.031 1.031 PCC VVI 18.88 8.39 15.56 8.41 Tap Operations 675 418 603 523
- 18. 18© 2014 Electric Power Research Institute, Inc. All rights reserved. Metric Improvement Based on Objective PV volt/var volt/watt power factor PCC Mean Voltage 1.031 1.027 1.031 1.031 PCC VVI 18.88 8.39 15.56 8.41 Tap Operations 675 418 603 523 PCC Mean Voltage 1.031 1.025 1.031 1.031 PCC VVI 18.88 30.21 15.56 14.18 Tap Operations 675 1727 603 485 PCC Mean Voltage 1.031 1.033 1.032 1.031 PCC VVI 18.88 6.02 9.92 8.41 Tap Operations 675 437 533 523 PCC Mean Voltage 1.031 1.034 1.032 1.032 PCC VVI 18.88 6.60 9.92 9.599 Tap Operations 675 401 533 485 Obj 4 Overall Obj 1 Reduce Overvoltage Obj 2 Improve Efficiency Obj 3 Reduce Taps
- 19. 19© 2014 Electric Power Research Institute, Inc. All rights reserved. Demo Site J1 Best Curves • Best volt/var and volt/watt curves shown for each objective • Each objective optimized with different curve characteristics Objective 1 Objective 2 Objective 3 Objective 4 Best Power Factor Setting Objective 1 2 3 4 power factor 0.90 0.97 0.94 0.97
- 20. 20© 2014 Electric Power Research Institute, Inc. All rights reserved. Impact of Load Level on Best Settings Peak Load Offpeak Load
- 21. 21© 2014 Electric Power Research Institute, Inc. All rights reserved. Sample Day – Comparing “Best” Setting Responses Offpeak day, highly variable solar 1.02 1.025 1.03 1.035 1.04 1.045 1.05 0 5 10 15 20 per‐unit voltage hour PCC Voltage 0 10 20 30 40 50 60 70 80 0 5 10 15 20 # hour Tap Operations Lesson Learned Power factor and proper volt/var settings can be effective
- 22. 22© 2014 Electric Power Research Institute, Inc. All rights reserved. Demo Site E1: Objectives & Metrics Objective Metric Weight 1. Reduce voltage flicker/voltage variations 100 2. Reduce losses 100 3. Combined 1 and 2 50/50
- 23. 23© 2014 Electric Power Research Institute, Inc. All rights reserved. Sample Plots 0 5 10 15 20 25 30 0.975 0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 0 5 10 15 20 25 30 0.985 0.99 0.995 1 1.005 1.01 1.015 0 5 10 15 20 25 30 0.975 0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 Clear Day Overcast day Highly variable day PCCvoltage hour PCCvoltage hour PCCvoltage hour
- 24. 24© 2014 Electric Power Research Institute, Inc. All rights reserved. Demo Site E1 Best Curves • 3 best volt/var and volt/watt curves shown for each objective • Each objective optimized with different curve characteristics Objective 1 Objective 2 Objective 3 Best Power Factor Setting Objective 1 2 3 power factor 0.92 0.99 0.93
- 25. 25© 2014 Electric Power Research Institute, Inc. All rights reserved. Metric Improvement Based on Objective PV volt/var volt/watt power factor PCC VVI 9.1787 7.4029 8.9498 6.8539 Losses (kWh) 3079 3038 3078 3124 PCC VVI 9.1787 7.0951 8.9498 6.8480 Losses (kWh) 3079 3082 3078 3129 PCC VVI 9.1787 7.4029 8.9498 8.1283 Losses (kWh) 3079 3038 3078 3091 Obj 3 Obj 1 Obj 2 power factor does not reduce losses
- 26. 26© 2014 Electric Power Research Institute, Inc. All rights reserved. Demo Site H1: Objectives & Metrics Objective Metric Weight 1. Flatter voltage and improved customer efficiency 50 50 2. Reduced LTC tap changes 100 3. Combined 1 and 2 25/25/50
- 27. 27© 2014 Electric Power Research Institute, Inc. All rights reserved. Demo Site H1: Best Curves • 1 best volt/var and volt/watt curves shown for each objective • Each objective optimized with different curve characteristics Best Power Factor Setting Objective 1 2 3 power factor 0.92 0.96 0.96 Objective 1 Objective 2 Objective 3
- 28. 28© 2014 Electric Power Research Institute, Inc. All rights reserved. Metric Improvement Based on Objective PV volt/var volt/watt power factor PCC Mean Voltage 1.008 1.008 1.008 1.007 PCC VVI 11.5798 10.7196 11.0827 8.0583 Tap Operations 54 53 54 55 PCC Mean Voltage 1.008 1.006 1.008 1.007 PCC VVI 11.5798 11.0533 11.0827 8.0391 Tap Operations 54 57 54 55 PCC Mean Voltage 1.008 1.009 1.008 1.007 PCC VVI 11.5798 9.4340 10.1347 8.0583 Tap Operations 54 50 53 55 Obj 3 Obj 1 Obj 2
- 29. 29© 2014 Electric Power Research Institute, Inc. All rights reserved. Summary • Overall “best” setting depends upon objective – Improve voltage – Increase efficiency – Regulator operations – Increase hosting • Preliminary analysis indicates trends in recommended settings can be found • Caution: Minor changes in settings (volt/var) can have significantly different impacts • Less “aggressive” settings work – Less risk, less potential benefit (e.g., increasing hosting capacity) • Results shown today are based upon site-specific conditions • Future work for determining recommended settings – Other locations – Other feeders – Combined inverters
- 30. Jeff Smith, Huijuan Li EPRI EPRI Smart Inverter Workshop, Santa Clara, CA 5/7/2014 Potential Interaction Between Smart Inverters
- 31. 31© 2014 Electric Power Research Institute, Inc. All rights reserved. Overview Objective Approach Evaluating potential inverter interaction Investigate possible inverter interaction resulting from smart inverter control on multiple PV systems Time-domain analysis in Matlab/Simulink to investigate possible inverter interaction
- 32. 32© 2014 Electric Power Research Institute, Inc. All rights reserved. Studied System PV1: 400 kW, 475 kVA PV2: 1000 kW, 1235 kVA 20 s simulation widow 1.2 Mvar Cap is switched on at 10 s, which causes voltage rise
- 33. 33© 2014 Electric Power Research Institute, Inc. All rights reserved. Interactions Between the Two Inverters Single PV providing vars Both PVs providing vars 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) V1 V2 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) V1 V2 Oscillations Observed !
- 34. 34© 2014 Electric Power Research Institute, Inc. All rights reserved. What May Impact Var Control Var control flow VoltVar Curve Inverter Averaging window V Reference Q Average V Q PI controller (Kp Ki) Switching Command Factors may impact var control: • Volt-var curve parameters • PI controller parameters: Kp and Ki • Voltage average window length
- 35. 35© 2014 Electric Power Research Institute, Inc. All rights reserved. Impact of Volt-var Parameters Volt/var 1 Volt/var 2 % AvailableVars voltage (pu) ‐100 Capacitive 100 Inductive 0.95 1.05 % AvailableVars voltage (pu) ‐100 Capacitive 100 Inductive 0.99 1.01 Controller parameters for both cases: Kp Ki 0.3 3
- 36. 36© 2014 Electric Power Research Institute, Inc. All rights reserved. 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) Volt/var 1 Voltages 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) V1 V1 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) V2 V2 Volt/var 2 High ratio may cause oscillations
- 37. 37© 2014 Electric Power Research Institute, Inc. All rights reserved. Volt/var 1 4 6 8 10 12 14 16 18 20 -100 -80 -60 -40 -20 0 20 40 60 80 100 Time (s) %ofAvailableVar Actual var Var reference Vars Var 1 Var 1 Var 2 Var 2 4 6 8 10 12 14 16 18 20 -100 -80 -60 -40 -20 0 20 40 60 80 100 Time(s) %AvailableVars Actual Var Var reference 4 6 8 10 12 14 16 18 20 -100 -80 -60 -40 -20 0 20 40 60 80 100 Time(s) %AvailableVars Actual var Var reference 4 6 8 10 12 14 16 18 20 -100 -80 -60 -40 -20 0 20 40 60 80 100 Time (s) %AvailableVars Actula var Var reference Volt/var 2
- 38. 38© 2014 Electric Power Research Institute, Inc. All rights reserved. Impact of Controller Parameters Volt/var 2 Volt/var 2: slower response % AvailableVars voltage (pu) ‐100 Capacitive 100 Inductive 0.99 1.01 Controller parameters for case 1: Kp Ki 0.3 3 % AvailableVars voltage (pu) ‐100 Capacitive 100 Inductive 0.99 1.01 Kp Ki 0.1 1 Controller parameters for case 2:
- 39. 39© 2014 Electric Power Research Institute, Inc. All rights reserved. 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) Volt/var 2 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu)Voltages V1 V1 V2 V2 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) Volt/var 2: slower response 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) Smaller control parameters, which means smaller adjustments at each step, reduce the oscillations
- 40. 40© 2014 Electric Power Research Institute, Inc. All rights reserved. 4 6 8 10 12 14 16 18 20 -100 -80 -60 -40 -20 0 20 40 60 80 100 Time(s) %AvailableVars Actual var Var reference Volt/var 2 4 6 8 10 12 14 16 18 20 -100 -80 -60 -40 -20 0 20 40 60 80 100 Time(s) %AvailableVars Actual Var Var reference 4 6 8 10 12 14 16 18 20 -100 -80 -60 -40 -20 0 20 40 60 80 100 Time(s) %AvailableVars Actual var Var reference Vars Var 1 Var 1 Var 2 Var 2 4 6 8 10 12 14 16 18 20 -100 -80 -60 -40 -20 0 20 40 60 80 100 Time(s) %AvailableVars Actual var Var reference Volt/var 2: slower response
- 41. 41© 2014 Electric Power Research Institute, Inc. All rights reserved. Impact of Window Length of Averaging Voltage Volt/var 2 Volt/var 2: larger avg window % AvailableVars voltage (pu) ‐100 Capacitive 100 Inductive 0.99 1.01 % AvailableVars voltage (pu) ‐100 Capacitive 100 Inductive 0.99 1.01 Controller parameters for both cases: Kp Ki 0.3 3 Length of average window= 0.05s Length of average window= 1 s
- 42. 42© 2014 Electric Power Research Institute, Inc. All rights reserved. Volt/var 2 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu)Voltages V1 V1 V2 V2 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) Volt/var 2: larger avg window 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) 4 6 8 10 12 14 16 18 20 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02 1.03 1.04 1.05 Time(s) Voltage(pu) Longer voltage averaging window increases oscillation magnitude, but improved dampening occurs
- 43. 43© 2014 Electric Power Research Institute, Inc. All rights reserved. Volt/var 2 4 6 8 10 12 14 16 18 20 -100 -80 -60 -40 -20 0 20 40 60 80 100 Time(s) %AvailableVars Actual Var Var reference 4 6 8 10 12 14 16 18 20 -100 -80 -60 -40 -20 0 20 40 60 80 100 Time(s) %AvailableVars Actual var Var reference Vars Var 1 Var 1 Var 2 Var 2 Volt/var 2: slower avg window 4 6 8 10 12 14 16 18 20 -100 -80 -60 -40 -20 0 20 40 60 80 100 Time(s) %AvailableVars Actual var Var reference 4 6 8 10 12 14 16 18 20 -100 -80 -60 -40 -20 0 20 40 60 80 100 Actual var Var reference
- 44. 44© 2014 Electric Power Research Institute, Inc. All rights reserved. Conclusions • Interactions exist between the two close by inverters • High ratio may cause oscillations • Smaller adjustments at each step as a result of smaller control parameters reduces oscillations • Longer voltage averaging window increases magnitude of oscillations, although dampening does occur