Skip to content
Chetan Chaudhari edited this page May 20, 2019 · 9 revisions

Welcome to the PVMismatch wiki!

Please use these pages to engage in discussion about PVMismatch.

Talks

Here are some presentations that use PVMismatch.

  1. A Fast Parameterized Model for Predicting PV System Performance under Partial Shade Conditions by Bennet Meyers at 2016 IEEE 43rd PVSC in Portland, Oregon

  2. PVMismatch Python Package at 8th PV Performance Modeling Workshop in 2017

  3. Accurate Modeling of Partially Shaded PV Arrays by Bennet Meyers at 2017 IEEE44th PVSC in Washington, DC

  4. Quantification of System-Level Mismatch Losses using PVMismatch by Chetan Chaudari in 2018 at WCPEC-7

  5. A Simplified Method to Approximation Bifacial System Mismatch Losses by Ronda Bailey at NREL 2018 bifiPV 2018 workshop in Denver

  6. The subtle art of bifacial perfromance modeling for fixed and tracking systems at 2019 PV Systems Symposium (PVPMC 12th)- by Silvana Ayala (NREL)

  7. Performance modeling of MLPEs at 2019 PV systems symposium (PVPMC 12th) - by Chetan Chaudhari (SunPower)

Roadmap

The most pressing items on the PVMismatch roadmap are as follows:

  1. Integration with PVLIB-Python. The key to this will be creating an abstract PVcell so that backend can be set by the user. This would allow the PVLIB-Python 5-parameter model to be used instead of the current 2-cell model already in PVMismatch.
  2. Fixing bugs and adding features. There are a few bugs and many feature requests in PVMismatch that aren't blockers, just a PITA.
  3. Even more speed improvements. There were massive speed improvements in v0.3.1 thanks to @bmeyers. But we can do even more, see #86 and #60.

Tips and Tricks

The following tips may help you avoid any known issues and eek out some more performance.

  1. Create a new pvsys = pvsystem.PVsystem() to set all cells to a new ambient irradiance instead of pvsys.setSuns(). As system size grows, setSuns() is slower and slower, but creating a new system is always about 0.5[s] thanks to cheap copies.