Short-Term Load Forecasting Method Based on Deep Reinforcement Learning for Smart Grid

Table 2

Parameter configuration of forecasting network.

Configuration	Parameter
(a) DNN network
Number of hidden layers	20
Number of neurons in layer 1–5 network	8192
Number of neurons in layer 6–10 network	4096
Number of neurons in layer 11–15 network	2048
Number of neurons in layer 16–20 network	1080
(b) Chaotic particle swarm optimization
Population number	15
Maximum number of iterations	35
Acceleration constant	0.26
Acceleration constant	0.48