search

NREL / reV

Renewable Energy Potential (reV) Model
https://nrel.github.io/reV/
BSD 3-Clause "New" or "Revised" License
107 stars 24 forks source link

reV generation failing for some but not all chunks when run on Himawari v7 data #378

Closed EvanRosenlieb closed 2 years ago

EvanRosenlieb commented 2 years ago

Bug Description When running reV Gen for a subset of Himawari v7 points, chunks 02 and 03 for all years always return mean_cf of 0 for all points. All other chunks return valid values. This results in a strip of longitudes for which there are 0 CF's:

Screen Shot 2022-08-10 at 5 13 52 PM

These chunks are returning an error in stdout:

Full Traceback Full traceback is in stdout of failed chunk, example here:

error.txt

To Reproduce Steps to reproduce the problem behavior Run reV gen using the config_gen.json in this directory: /shared-projects/rev/projects/seasia_fpv/

Also uploading here (as text files):

config_gen.txt

samconfig_pv_fixtilt_south.txt samconfig_pv_fixtilt_north.txt

Expected behavior ReV gen to complete for all chunks without error

System (please complete the following information): Eagle, using shared reV env.

Additional context Have not been able to find anything odd in the meta data for these points, or in the timeseries resource values for these points (at the very least, spot checking some points within these chunks doesn't show odd values for dni, dhi, windspeed, air temperature)

Charge code WFED.11556.03.02.01

grantbuster commented 2 years ago

The weird himawari meta data with [brackets] in country names was being interpreted as a list input and failed when getting passed into json decode. Added a hot fix here: https://github.com/NREL/reV/commit/13525b8802e526dca09cd8ee66784a9f3e01eb60