SOLAR PV SIMULATIONS FOR CLEAR SKY CONDITIONS

Hello! I know that SAM is primarily for simulations under real-world weather conditions.

But are there any weather files available representing clear sky conditions for use in SAM simulations? 

I have some clients who have noted, like myself, the variation in annual solar output with one-axis tracking versus the annual output using the optimum fixed tilt angle. If you do simulations for a range of latitudes and any given longitude, the result - as you would expect - is a correlation curve with significant scatter on account of the cloud cover varying according to both latitude and longitude. On the other hand, if one were to do the same set of simulations for clear sky conditions, then the only variable is latitude and the correlation curves will be smooth and the curve for the ratio (annual one-axis output / annual output with optimum fixed tilt angle) will also be smooth.  If I can produce these smooth correlation curves to show to clients, it's far easier to explain things to them when you can demonstrate that the scatter based on simulations and real-world weather is due to the cloud cover being different from place to place, and depends on both latitude and longitude.

Robert T. Chisholm, Ottawa, Canada

Hi Robert,

You can download weather files with clear sky data from the NREL National Solar Radiation Database (NSRDB). SAM does not include that data when you download files from the Location and Resource input page, but if you download files direclty from the NSRDB Data Viewer at maps.nrel.gov/nsrdb-viewer/ , you can include the clear sky data in your request.





If you use that approach, you will have to use a text editor or spreadsheet software to edit the column headings and trick SAM into using the clear sky data instead of the regular DNI and DHI data: Rename the "Clearsky DNI" column heading to "DNI" and "Clearsky DHI" to "DHI" and also rename the original DNI and DHI column headings to something like "DNI original" and "DHI original" to prevent SAM from reading those columns. SAM ignores any columns of data in the file that it doesn't recognize so you don't need to remove the data itself.

Best regards,
Paul.

OK Paul - many thanks for the info; I'll try this out and let you know what happens.

One question at this point, thought: can one use this procedure, or a similar one,  for latitudes higher than 60 deg. North and particularly in Canada? This factors will apply to most of the project locations that I'm interested in. 

Such locations - from what I've seen to date - currently lie outside the scope of the NSRDB database. I have been getting "real world" weather files in .smw format for these locations using the getmerra2 and makeweatherfiles  software from Sourceforge, going all the way to the North Pole. If I can do something with these .smw files similar to what you recommended using with the NSRDB weather data files, that would be a big help.

Thanks once again!

Sincerely, Robert T. Chisholm

Hi Robert,

The approach I describe only works for locations within the area covered by the NSRDB.

We do have a Wish List item to have SAM calculate report the clear sky irradiance values, but that work is not part of a funded project yet: github.com/NREL/ssc/issues/151 . I made a note of your request there to help us as we prioritize our task list.

Best regards,
Paul.

As of SAM 2020.11.29 r2 (not yet released as of 5/6/2021), SAM will download clear sky ("clearsky" for web searches) DNI, DHI, and GHI data from the NSRDB, so the weather file will contain the data to use for comparisons. We will need to make more modifications to SAM to display the data in simulation results, but for now, at least the data will be available without having to download files directly from the NSRDB website.

This change is documented in pull request 609 on the SAM GitHub repository:

github.com/NREL/SAM/pull/609

Best regards,
Paul.

Thanks for following up on this, Paul!