06 2017.05.09 freeman 8th pvpmc iec 61853 presentation
0 likes968 views
This document summarizes improvements made to the System Advisor Model (SAM) for predicting photovoltaic module performance based on IEC-61853 test data. A new model was developed that estimates module parameters at any test condition or operating condition, resulting in an 88% reduction in error compared to the previous 6 parameter model. The updated model and its ability to more accurately reproduce module test data from the improved parameter estimation technique will be available in SAM around July.
06 2017.05.09 freeman 8th pvpmc iec 61853 presentation
- 1. Significant Improvement in Module Performance Prediction Accuracy Based on IEC-61853 Data Janine Freeman 8th PV Performance Modeling Workshop Santa Ana Pueblo, New Mexico May 9, 2017
- 2. 2 Irradiance Transposition using Isotropic, HDKR, or Perez Measured plane of array (POA) input Shading Irregular obstruction shading from 3D scene Self-shading for regularly spaced rows External input from SunEye, Solar Pathfinder Snow cover loss model Module Simple efficiency model Single diode model (CEC database or datasheet) Extended single diode model (for IEC-61853 tests) Sandia PV Array Performance Model Inverter Sandia/CEC grid-tied inverter model Datasheet part-load efficiency curve System Sizing wizard or electrical layout Multiple subarrays Fixed, 1 axis, backtracking, azimuth axis, 2 axis Battery storage Degradation Extrapolated single year Lifetime simulation of all years Simulation 1 minute to 1 hour time steps Detailed photovoltaic model ֍ System Advisor Model
- 3. 3 Background System Advisor Model 1. Equivalent circuit models 2. Point models 3. “Data” models
- 4. 4 Background System Advisor Model Single diode model Great at STC, but what about all these conditions?
- 5. 5 Background System Advisor Model Original model implementation in SAM: “Procedure for applying IEC-61853 test data to a single diode model”, Dobos, PVSC 2014 Based only on thin film data! Curve fitting errors sometimes reduced ability of the model to accurately reproduce test data
- 6. 6 Estimating Parameters at Test Conditions System Advisor Model
- 7. 7 Estimating Parameters at Any Condition System Advisor Model
- 8. 8 Estimating Parameters at Any Condition System Advisor Model
- 9. 9 Map Boundaries System Advisor Model If desired operating condition is… • sufficiently near edge of convex hull defined by test points, nearest known point used • Too far away, model reverts to standard single diode model equations ֍
- 10. 10 Comparison with 6 Parameter Model System Advisor Model
- 11. 11 Comparison with 6 Parameter Model System Advisor Model 88% reduction in error on average across 20 modules
- 12. 12 Upcoming Work System Advisor Model Further validation compared to measured data using MPERT data
- 13. 13 • Desktop Application • Built-in Scripting Language • Software Development Kit (SDK) o C/C++ o Matlab o Python o C# o Java • Upcoming: SAM open-sourcing (more on that later!) Where can I get the new model? System Advisor Model Dobos, A.; Freeman, J.: Significant Improvement in PV Module Performance Prediction Accuracy Using a New Model Based on IEC-61853 Data, forthcoming New Model Implementation Release Expected ~July
- 14. Our team Janine Freeman Project lead, PV and wind models Nick DiOrio Software architecture, Battery storage models Steve Janzou Programming, financial structures (subcontractor) Paul Gilman User support and documentation (subcontractor) Ty Neises Concentrating solar power models Mike Wagner Concentrating solar power models Nate Blair Emeritus lead, financials, costs, systems http://sam.nrel.gov System Advisor Model