One-Step Two-Critic Deep Reinforcement Learning for Inverter-based Volt-Var Control in Active Distribution Networks.
Qiong LiuYe GuoLirong DengHaotian LiuDongyu LiHongbin SunWenqi HuangPublished in: CoRR (2022)
Keyphrases
- distribution networks
- reinforcement learning
- control problems
- optimal control
- temporal difference
- function approximation
- actor critic
- post processing
- robot control
- reinforcement learning algorithms
- robotic control
- induction motor
- adaptive control
- model free
- control algorithm
- lead time
- action selection
- control strategy
- control policy
- state space
- machine learning
- control strategies
- policy gradient
- control system
- multi agent
- learning algorithm
- control method