P50/P90 Analysis - Weather data base files

3 posts / 0 new
Last post
P50/P90 Analysis - Weather data base files

We are performing a P50/P90 analysis for one of the CSP plants near Jodhpur in Rajasthan, India. We have considered a nearby location Jaipur for carrying out the analysis. We note that the available weather data is for 7 years for this location.

Further, while carrying out the analysis we note that the P50/P90 results are not being tabulated. The error given is that the data is insufficient to carry out the P90 analysis.

Can you please suggest how to handle this limited input data using SAM to carry out the P50/P90 analysis.

Paul Gilman


SAM's P50/P90 analysis feature requires at least 15 years of data, which we consider to be the absolute minimum number of years for which it makes sense to do this kind of statistical analysis.

If you want to calculate exceedance probabilities over a period of fewer than 15 years, you may want to consider doing the calculation outside of SAM, or writing a SamUL script to do the calculation in SAM. With only 7 years of values, the calculations should be straightforward to make in a spreadsheet.

Best regards,

GeoModel Solar

Also worth mentioning is that if a P90 analysis is done for a location near Jodhpur, using a dataset for Jaipur (approx. 300 kms away), the P90 value has to be very conservative. This by no means would be appreciated by any project developer. Looking at some DNI maps, it appears that DNI in Jaipur is approx 15% lower than DNI in Jodhpur.

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer