• Home
  • Search
  • Deep Reinforcement Learning Optimizing for Geological CO2 Storage Considering Geomechanical and Leakage Risks
  • https://doi.org/10.56952/igs-2025-0143Copy DOI Icon

Deep Reinforcement Learning Optimizing for Geological CO2 Storage Considering Geomechanical and Leakage Risks

  • Nov 18, 2025
  • Gamze Erdogan Erten +4 more
Show More
  • Abstract
  • Literature Map
  • Similar Papers
Abstract

Optimizing well placement and injection rates to maximize CO2 storage while limiting leakage and geomechanical risks is a critical challenge for scaling up geological carbon storage (GCS). This study develops a deep reinforcement learning (DRL) framework in which a convolutional neural network (CNN) policy maps surrogate-predicted CO2 saturation and pressure fields to well locations and dynamic injection rates. Policy parameters are trained using Proximal Policy Optimization (PPO) with a multi-objective reward that promotes CO2 storage while penalizing pressure buildup and boundary leakage. High-fidelity surrogate models, trained on coupled flow-geomechanics simulations, enable efficient agent–environment interactions at reduced cost, supporting high-resolution evaluation across diverse geological realizations. Geological uncertainty is addressed through domain randomization, where 1,000 permeability realizations are embedded via multidimensional scaling and grouped into 20 clusters for streamlined training. The framework is applied to a conditional geostatistical model of the Aquistore storage site in Saskatchewan, Canada. Performance is benchmarked against fixed-rate injection schedules and the covariance matrix adaptation evolution strategy (CMA-ES). Results show that PPO consistently identifies superior well configurations and dynamic injection strategies, achieving greater stored volumes while maintaining operational safety across all validation samples.

Similar Papers
  • Research Article
  • Citations6

Optimization of underground mining production layouts considering geological uncertainty using deep reinforcement learning

  • Oct 23, 2024
  • Engineering Applications of Artificial Intelligence
  • Roberto Noriega +1
  • Research Article
  • Citations1

Optimizing the user personalized recommendation system of new media short videos by using machine learning

  • May 08, 2025
  • Journal of Computational Methods in Sciences and Engineering
  • Danqing Zhu
  • Research Article
  • Citations3

Geophysical control of geologic carbon storage using deep reinforcement learning: Sensitivity to multigeophysical noise and to the uncertainty of digital twins

  • Jun 13, 2025
  • GEOPHYSICS
  • Kyubo Noh +1
  • Research Article
  • Citations33

Improving the interpretability of deep reinforcement learning in urban drainage system operation

  • Nov 23, 2023
  • Water Research
  • Wenchong Tian +4
  • PDF
  • Research Article
  • Citations8

A K-means Supported Reinforcement Learning Framework to Multi-dimensional Knapsack

  • Feb 15, 2024
  • Journal of Global Optimization
  • Sabah Bushaj +1
  • Dissertation

Machine learning for signal integrity and power integrity analysis

  • Jan 01, 2025
  • Weiyang Miao
  • Conference Article
  • Citations2

Washing Optimization Method Based on Deep Reinforcement Learning for Fully Programmable Valve Array Biochips

  • Jun 12, 2024
  • Junqian Huang +3
  • Conference Article
  • Citations4

Learning Similar Tasks Based On PPO By Transferring Trajectory

  • May 01, 2019
  • An Guo +2
  • Research Article
  • Citations5

Automating the optimization of proton PBS treatment planning for head and neck cancers using policy gradient-based deep reinforcement learning.

  • Jan 31, 2025
  • Medical physics
  • Qingqing Wang +1
  • Conference Article
  • Citations1

Optimizing Oil Field Development with Multi-Action Decision-Making in Deep Reinforcement Learning

  • Oct 13, 2025
  • Y Falola +3
  • Conference Article

Cooperative Work Simulation of Construction Equipment using Multi-Agent Deep Reinforcement Learning

  • Nov 28, 2025
  • Guozhong Cheng +4
  • Preprint Article

Closing the Gap to the Oracle: Benchmarking Domain-Informed Deep Reinforcement Learning for Deficit Irrigation against Perfect Foresight

  • Mar 14, 2026
  • Niels Schuetze +2
  • Research Article
  • Citations14

A trustworthy reinforcement learning framework for autonomous control of a large-scale complex heating system: Simulation and field implementation

  • Nov 08, 2024
  • Applied Energy
  • Amirreza Heidari +3
  • Video Transcripts

Deep reinforcement learning for large-scale epidemic control

  • May 07, 2020
  • Underline Science Inc.
  • Timothy Verstraeten +6
  • Research Article
  • Citations21

Bayesian Optimization for Field-Scale Geological Carbon Storage

  • Nov 01, 2022
  • Engineering
  • Xueying Lu +4
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.