DeepThermal: Combustion Optimization for Thermal Power Generating Units Using Offline Reinforcement Learning

This is the code of the paper DeepThermal: Combustion Optimization for Thermal Power Generating Units Using Offline Reinforcement Learning accepted at AAAI'2022. The paper can be found here.

Usage

The code of combustion simulator is in Simulator/simrnn_model.py, the code of model-based offline RL framework, MORE, is in RL/primal_dual_ddpg.py.

Bibtex

@inproceedings{zhan2022deepthermal,
  title={Deepthermal: Combustion optimization for thermal power generating units using offline reinforcement learning},
  author={Zhan, Xianyuan and Xu, Haoran and Zhang, Yue and Zhu, Xiangyu and Yin, Honglei and Zheng, Yu},
  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
  pages={4680--4688},
  year={2022}
}