Actor-critic reinforcement learning to estimate the optimal operating conditions of the hydrocracking process.
Dong-Hoon OhDerrick AdamsNguyen Dat VoDela Quarme GbadagoChang-Ha LeeMin OhPublished in: Comput. Chem. Eng. (2021)
Keyphrases
- operating conditions
- reinforcement learning
- actor critic
- optimal control
- control strategy
- dynamic programming
- fault diagnosis
- average reward
- function approximation
- policy gradient
- steady state
- genetic algorithm
- decision trees
- approximate dynamic programming
- supervised learning
- fuzzy sets
- model free
- reinforcement learning algorithms
- artificial intelligence