PVWatts single year to lifetime outputs

[cpaulgilman]

Describe the bug

Some PVWatts outputs in SAM 2026.7.3 are different from SAM 2025.4.16, possibly due to change from single year to lifetime simulations.

To Reproduce

1. Create a default PVWatts / Single Owner case in SAM 2025.4.16.
2. Run a simulation and note annual output and NPV in metrics table, and note AC inverter output power in hourly data outputs.
3. Save the file.
4. Open it in SAM 2026.7.3 and compare outputs.

Expected behavior

AC inverter output power in SAM 2026.7.3 should be the same as SAM 2026.4.16 for a case created in SAM 2025.4.16.

Financial outputs should be the same or very close.

Operating System and Version

SAM 2026.7.3 for Windows

[brtietz]

@cpaulgilman - I disagree with the expected behavior here. 2025.4.16 used compounding AC degradation (degradation set to 5% to make it obvious):

Whereas 2026.7.3 uses linear DC degradation (a better fit to the literature as per NatLabRockies/ssc#503)

While the change in the compute module dates back a few years, we started running it in lifetime mode in the UI in 2026.7.3.

What happens if degradation is set to 0? Is that useful as a debugging case?

[cpaulgilman]

@brtietz Good point! Degradation explains the difference in NPV and other financial metrics, and yes, setting degradation to zero is a good test case. I'll revise the description in the release notes to make the description of degradation change less subtle.

I think there's still an issue with time series output variables like AC inverter output power.

[brtietz]

@cpaulgilman Another piece of evidence here - some ac energy single value outputs still display year 1 values, while others display year 25 values:

Is it possible some of what we're seeing in the buggy outputs are degraded values?

[cpaulgilman]

Time series (hourly) AC inverter output power for "test" case from 2026.7.3 is on the order of 0.5 of values in 2025.4.16 r2. For "default" case, difference is about 0.88. Default case has AC degradation = 0.5%/yr, test case has AC degradation = 2%/yr. See pvwatts-so-2025-4-16.zip.

Note that converting PVWatts-Battery to detailed battery (#2113) involved changing PVWatts degradation from AC degradation to DC degradation in SAM 2026.7.3. This is not documented in release notes.

SAM 2025.4.16 test case:

SAM 2026.7.3 test case: