Research: Blind photovoltaic modeling intercomparison: A multidimensional data analysis and lessons learned

[pdvjakgmail.com]

Hello,

One of my clients came across the referenced research document (of which Mr. Bruno Wittmer is co-writer, I believe) and is worried about '14% variation' mentioned in 

And most recently, Vogt et al.12 conducted an intercomparison of energy rating calculations per IEC 61853-313 with nine European institutes. Energy rating differences of 14% were found in the first blind comparison round. It took five rounds of calculations—and discussions among the participants—for the nine participants' calculations to agree within 0.1%. Ultimately, Vogt et al.12 demonstrated how user-induced variability can be reduced when modelers have clear procedures for implementing key steps of the PV model chain.

Client suggests "My only ask is that you consider which PVSyst model and use your expertise to tune the parameters as appropriate."

Under what conditions would 'tuning' which parameters be required or advisable?

(Personally I have good confidence in PVSyst settings, but am obliged to ask the question)

(Tried to upload the PDF of the paper in question, but that failed with a "-200" error message?)

Thank you for any insights...

Pieter F. de Vries - Principia.solar