Optimal-Energy-System-Scheduling-Combining-Mixed-Integer-Programming-and-Deep-Reinforcement-Learning
Optimal-Energy-System-Scheduling-Combining-Mixed-Integer-Programming-and-Deep-Reinforcement-Learning copied to clipboard
State space issue
Hi, Hou.
Thanks for sharing. Just one question regarding the state space limits. It seems that the net_load could be negative. In that case, shouldn't the normalized state space defined as [-1,1]?
Thanks.
This doesn't impact the performance according to my experience, state normalization and denormalization is to stable the training by scaling the state into a magnitude that agents can easily handle. But you can try to do some ablation cases using my framework. I did several methods for normalization when running the model, but didnt see any big performance derivations.