IBM / rl-testbed-for-energyplus

Reinforcement Learning Testbed for Power Consumption Optimization using EnergyPlus
MIT License
177 stars 74 forks source link

Handle output variable name change since E+ 9.4.0 #94

Closed antoine-galataud closed 2 years ago

antoine-galataud commented 2 years ago

Signed-off-by: Antoine Galataud antoine@foobot.io

Fixes #89

Handle Whole Building:Facility Total * variables renamed.

antoine-galataud commented 2 years ago

tested with E+ 9.6.0 and 22.1.0, works ok:

root@57acb82563a6:~/rl-testbed-for-energyplus# python3 -m baselines_energyplus.common.plot_energyplus
energyplus_locate_log_dir: /root/eplog/openai-2022-05-19-09-04-40-711667
Logging to /tmp/openai-2022-05-19-09-18-06-867003
Monitor: filename=/tmp/openai-2022-05-19-09-18-06-867003
energyplus_plot.plot log=/root/eplog/openai-2022-05-19-09-04-40-711667
episode 1
read_episode: file=/root/eplog/openai-2022-05-19-09-04-40-711667/output/episode-00000001/eplusout.csv.gz
Reward                    ave=-5.60, min=-14.05, max= 0.60, std= 1.87
westzone_temp             ave=41.98, min=16.48, max=104.64, std=11.28
eastzone_temp             ave=43.59, min=17.24, max=95.26, std=11.45
Power consumption         ave=187,476.49, min=66,792.74, max=309,058.14, std=36,946.41
pue                       ave= 1.07, min= 1.01, max= 1.72, std= 0.07