Battery Charging Bug

[dguittet]

The calculate_current_for_power_kw function reduces the battery's target power and current to what is available given SOC. Fix the bug in the charging portion where the reduced power and current should have been negative.

Fix #1361

[brtietz]

Thanks for looking into this! A few requests:

1. Can you add a test that establishes this sign convention explicitly? There are a few tests in lib_resilience_test that are close, but I'm getting lost with multiple negative sign conventions. Bonus points for establishing the convention in the discharging function as well.
2. Can you review other uses of the function to ensure the sign convention is correct? I'm seeing some additional inconsistencies in the below.

[dguittet]

Thanks Brian for the good catch. Turns out it was just a few edge cases here and there where calculate_max_discharge_kw and calculate_max_charge_kw could return values that were outside of their expected bounds so I added limit checks and removed the erroneous sign change in calculate_current_for_power_kw

[coveralls]

Coverage Report for PR #1362

Coverage decreased (-0.006%) to 56.284%

Diff Coverage: No coverable lines changed

Coverage Regressions

16 previously-covered lines in 3 files lost coverage.

File	Lines Lost	Coverage
ssc/shared/lib_battery_powerflow.cpp	4	95.94%
ssc/shared/lib_geothermal.cpp	5	65.99%
ssc/shared/lib_battery.cpp	7	89.87%

---

Change from base Build 23860928407:	-0.006%
Covered Lines:	68172
Relevant Lines:	121122

---

💛 - Coveralls

[brtietz]

Thanks for incorporating the fixes!