Really Low Summer Generation for a California Solar Farm?

Hey All,I am validating the usage of SAM (via reV) and I am getting a strange result when looking at the 12x24s of actual generation compared to the modeled generation that is output from SAM. I am using the pvwatts7 module and I have the exact latitude and longitude of the farm. Here is a code snippet of how I am calling SAM within reV in python 3.9.18 using HSDS
And here is my json of parameters: 
Attached is a graph of the results compared to the actuals in which you can see that months 6-8 have a much lower capacity factor than the actuals and a much lower capacity than the model predicted in February! Please let me know if there is something I can do with my parameters or the model code to fix this aberration. Thank you!Luke

Hi Luke,

Those plots are hard to read, so I couldn't see the exact results you describe.

The first place to look would be in the weather data. The main driver for output of the array is the direct normal irradiance (DNI) so you might see what that data looks like in those months of lower-than-expected capacity factor. Ambient temperature is also a factor: Months of especially high ambient temperatures may have lower capacity factors than expected.

Best regards,
Paul.

Hey Paul,

Thanks for the response. Sorry if my plot was hard to read. Generally the story is summer generation for a solar farm modeled by SAM/REV is 30% lower than actuals: an unacceptable material difference. Is there a way to take temperature out of the equation? I assume it is a temperature effect and that we are not the first people so see this. It would make sense if there's a parameter to pass or a section to comment out of the code so that everyday is treated as cold and high-efficiency. Thanks again for the help.

-Luke

Hi Luke,

The code snippet you provide above runs pvwattsv7, which is an old version of PVWatts. The current version is PVWatts V8 ('pvwattsv8' compute module). When you run PVWatts from the SAM desktop application or the online PVWatts calculator ( pvwatts.nrel.gov ), you are running PVWatts V8.

PVWatts is different from SAM's Detailed PV model ('pvsamv1').

I just want to clarify this to be clear about what model we are discussing.

Comparing results from a model like SAM's Detailed PV model or PVWatts to measured data from a real system is tricky. Some of the short papers available on the PV Validation page of the SAM website at sam.nrel.gov/photovoltaic/pv-validation.html describe the steps required to make sure the measured data and model results are comparable.

It is not possible to turn off the cell temperature model in either PVWatts or the Detailed PV model.

Best regards,
Paul.