@mikofski Have you ever seen evidence that this change in shunt resistance is actually an irradiance phenomenon rather than a temperature phenomenon?

I have been fitting many curves in the mPERT outdoor datasets, and it's hard to say what the root cause of the change in shunt resistance is, because the temperature and irradiance are strongly correlated positively in these datasets.

Perhaps there are some IEC 61853-1 matrix measurements that clearly show the trend in the irradiance dimension but not in the temperature direction?

Both the PVSyst and DeSoto/CEC models have an irradiance dependence for shunt resistance although I don't know what the physical explanation is. Kyumin didn't discuss it at all in his presentations, only his proposed temperature dependence of series resistance due to metallization. I agree there is a strong correlation between temperature and irradiance, so maybe the PVSyst and DeSoto models use shunt resistance as a proxy for temperature? They also have band gap temperature correction (DeSoto) or gamma (diode ideality factor) temperature correction (PVSyst) so maybe not?

Those are both also single diode models, not two diode models like PVMismatch. I also hoped that we would discover some new algorithm that would use Isat2 to compensate for irradiance. Oh well.

Regardless, the relative error in Pmp increases as irradiance decreases for every temperature without some correction. @chetan I made some corrections, so I'll update this PR soon