CCS_E2CO-RL

Code for "Optimization of pressure management strategies for geological CO2 storage using surrogate model-based reinforcement learning".

under construction

step 1: run MSE2C_Ksteps.ipynb to construct E2CO proxy models (either one should work, maybe MSE2C_Ksteps_SC.ipynb work best)

step 2: run RL_SAC_train.ipynb to employ RL to optimize

Citation

If you find our research helpful, please consider citing us with：

@article{chen2024optimization,
  title={Optimization of pressure management strategies for geological CO2 storage using surrogate model-based reinforcement learning},
  author={Chen, Jungang and Gildin, Eduardo and Kompantsev, Georgy},
  journal={International Journal of Greenhouse Gas Control},
  volume={138},
  pages={104262},
  year={2024},
  publisher={Elsevier}
}