Ssc 1300 wind lifetime mode

[brtietz]

Pull Request Template

Description

Adds lifetime mode to windpower. In addition to the obvious for loops, required updating the data provider with a rewind function. This reads in the file at the start of the simulation and references arrays, more similar to PV. AI disclosure: Claude Sonnet assisted with the rewind function.

Note that the weather file mode for wind is slower by a linear factor of analysis period length. Open to changing that prior to merging if we prefer a different approach to applying degradation. This would impact year 1 vs lifetime outputs.

Fixes #1300

Corresponding branches and PRs:

SAM PR tbd. Develop of other repos.

Unit Test Impact:

New wind lifetime tests. No other changes expected

Checklist

• requires help revision and I added that label
• adds, removes, modifies, or deletes variables in existing compute modules
• adds a new compute module
• changes defaults
• I've tagged this PR to a milestone

Reminders- this section can be deleted

[Checking for PySAM Incompatible API Changes]
(https://github.com/NREL/SAM/wiki/PySAM-Incompatible-API-Changes-&-Regenerating-PySAM-Files).

[When do the PySAM files need to be regenerated?]
(https://github.com/NREL/SAM/wiki/PySAM-Incompatible-API-Changes-&-Regenerating-PySAM-Files#when-do-the-pysam-files-need-to-be-regenerated-via-export_config)

[coveralls]

Coverage Report for CI Build 25519283432

Coverage increased (+0.006%) to 56.281%

Details

• Coverage increased (+0.006%) from the base build.
• Patch coverage: No coverable lines changed in this PR.
• 236 coverage regressions across 6 files.

Uncovered Changes

No uncovered changes found.

Coverage Regressions

236 previously-covered lines in 6 files lost coverage.

File	Lines Losing Coverage	Coverage
ssc/shared/lib_pv_io_manager.cpp	88	83.95%
ssc/ssc/cmod_windpower.cpp	53	86.87%
ssc/shared/lib_windfile.cpp	48	62.07%
ssc/ssc/cmod_pvwattsv8.cpp	36	82.83%
ssc/shared/lib_battery.cpp	7	89.87%
ssc/shared/lib_battery_powerflow.cpp	4	95.94%

---

Coverage Stats

Relevant Lines:	121201
Covered Lines:	68213
Line Coverage:	56.28%
Coverage Strength:	3454450.87 hits per line

---

💛 - Coveralls

[cpaulgilman]

Degradation does not appear to be applied correctly.

Results for default Wind Power / Single Owner case with 5% AC degradation rate lifetime simulation:

Compare to SAM 2025.4.15 r1 single year simulation:

[brtietz]

Degradation does not appear to be applied correctly.

Thanks for flagging this. I grabbed the code from PVWatts, which applies the degradation linearly:

ssc/ssc/cmod_windpower.cpp

Line 378 in 2ffaf57

degradationFactor.push_back(1.0 - (ac_degradation[0] * y) / 100.0);

The code in the released version applies degradation exponentially:

ssc/ssc/cmod_singleowner.cpp

Line 1155 in 2ffaf57

for (i = 1; i <= nyears; i++) cf.at(CF_degradation, i) = pow((1.0 - degrad[0] / 100.0), i - 1);

Looks like I used the exponential convention for custom generation profile as well:

ssc/ssc/cmod_custom_generation.cpp

Line 138 in 2ffaf57

sys_degradation.push_back((ssc_number_t)pow((1.0 - (double)degrad[0] / 100.0), i));

Just to confirm/establish a convention here: DC degradation for PV is linear, but for all other models AC degradation is exponential. Is that what we want? (if so I'll want to adjust some version upgrade messaging around the PVWatts changes for this version)

[brtietz]

I pulled up some papers to help inform this decision. The curves in https://www.sciencedirect.com/science/article/pii/S0360544225017840 and https://papers.phmsociety.org/index.php/phme/article/download/3328/1939 look pretty linear. ATB assumes that O&M will counteract degradation, which is unhelpful here.

@janinefreeman is there an NLR wind person you'd recommend asking about this?

[janinefreeman]

I've messaged a wind person to ask for the right name and I'll send you a Teams message. It does seem weird to me that DC degradation would be linear but it also rings a bell, is there an old issue that might illuminate that decision somewhere?

[brtietz]

I've messaged a wind person to ask for the right name and I'll send you a Teams message. It does seem weird to me that DC degradation would be linear but it also rings a bell, is there an old issue that might illuminate that decision somewhere?

Thanks! See #503 for some of the PV background.

[cpaulgilman]

I've messaged a wind person to ask for the right name and I'll send you a Teams message. It does seem weird to me that DC degradation would be linear but it also rings a bell, is there an old issue that might illuminate that decision somewhere?

Thanks! See #503 for some of the PV background.

Regarding PV degradation, I did a quick search on the PVPMC Modeling Steps and pvlib websites and did not see mention of annual DC degradation there.

[brtietz]

https://www.sciencedirect.com/science/article/pii/S0960148113005727#fig9 is a good citation for linear degradation for wind. The recent changes add a page to explain linear versus compounding by model using AC degradation.

I've also updated startup.lk to get all of the hybrids in lifetime mode, which should solve the issues caught in NatLabRockies/SAM#2166

Outstanding issues include:

1. Version upgrade doesn't work for hybrids. In the 5 tech configuration, custom generation profile degradation makes it through the upgrade, but PVWatts and wind keep their defaults
2. The wind default degradation has some annoying floating point issues. It should be 0.64 but is just below that.

Any suggestions on those two prior to a more formal re-review?

[cpaulgilman]

I found a couple more instances of generic_degradation in SSC (none in SAM/SAM_api):

\ssc\test\input_cases\general_data\pvsmoothing_Generic_alloptions.json

• Line 11: "generic_degradation" : [ 0 ]

\ssc\ssc\cmod_generic_system.cpp

• Line 66: { SSC_INPUT, SSC_ARRAY, "generic_degradation", "Annual AC degradation", "%/year", "", "Lifetime", "system_use_lifetime_output=1", "", "" }
• Line 145: degrad = as_array("generic_degradation", &count_degrad);