Data-driven optimal control of wind turbines using reinforcement learning with function approximation.
Shenglin PengQianmei FengPublished in: Comput. Ind. Eng. (2023)
Keyphrases
- optimal control
- function approximation
- reinforcement learning
- wind turbine
- wind speed
- control problems
- dynamic programming
- temporal difference
- model free
- temporal difference learning
- feedback control
- state space
- function approximators
- reinforcement learning algorithms
- td learning
- power plant
- optimal policy
- multi agent
- markov decision processes
- approximate dynamic programming
- infinite horizon
- machine learning
- learning algorithm
- fault diagnosis
- learning process
- supervised learning
- actor critic
- transfer learning
- policy gradient
- control law
- reinforcement learning methods
- policy iteration
- action selection
- neural network
- control strategy
- data mining