Differences w/ SAM App, SDK PV Model

Hi,

I'm setting myself up to model some PV systems across the country, using the SDK to sweep over a number of points. Before launching off, I've been trying to get the SAM App output to match that of the SDK for the same configuration and have been unable to do so. I've isolated the issue down to the "Module loss" component. I found the mismatch for setups using both the CEC and 6par module models, and two different setups using the 6par model. I've checked my module inputs in all cases at least three times.

(1) Is there a way to "look inside" the "Module loss" output to diagnose where specifically the issue is?

(2) Any general tips for diagnosing where my issue is?

(3) Should I not expect the models to match perfectly?

Thanks,
Daniel

Dear Daniel,

Here are some tips that may help:

Make sure that you are using the same version of SSC in the SDK and SAM. To see the SSC version number in SAM, click Help, and then, in the Help window, click About. The latest version of SSC is 159.

You can generate a set of SSC inputs as an LK script from SAM by pressing the F5 key.

The SDKtool application that comes with the SDK is a useful way to do this kind of calibration. You can generate the list of input variables from SAM and then import that script into the SDKtool script editor. You can then either modify the script to run the pvsamv1 module directly in the script editor by adding run('pvsamv1') to the end of the script, or you can run it from the module browser by choosing the pvsamv1 module from the list at the bottom left of the page next to the Run button (first run the LK script to assign values to inputs). After you run a simulation, you can use the data container to see the values of inputs and outputs. The module browser, data container, and script editor are tabs in the SDKtool.

As for your questions:

1. No, unfortunately, there is not a way to break down the module loss value. It includes effects of cell temperature and incidence angle, but results of those equations are not reported separately. Also, each module model uses a different approach to model those effects.

2. Hopefully my suggestions above help.

3. As long as the inputs are identical and the SSC version is the same, you should expect the results to be the same. There may be some small differences caused by differences in rounding and number formatting between SAM user interface and your code.

Best regards,
Paul.