Uncertainty Analysis of CO2 Storage in Deep Saline Aquifers Using Machine Learning and Bayesian Optimization

Geological CO2 sequestration (GCS) has been proposed as an effective approach to mitigate carbon emissions in the atmosphere. Uncertainty and sensitivity analysis of fate dynamics storage are essential aspects large-scale reservoir simulations. This work presents a rigorous machine learning-assisted (ML) workflow for uncertainty global prediction deep saline aquifers. The comprises three main steps: first step concerns dataset generation, which we identify parameters impacting flow transport then determine their corresponding ranges distributions. training data samples generated by combining Latin Hypercube Sampling (LHS) technique with high-resolution second involves ML model development based on data-driven model, is map nonlinear relationship between input output interests from previous step. We show that using Bayesian optimization significantly accelerates tuning process hyper-parameters, vastly superior traditional trial–error analysis. In third step, performed Monte Carlo simulations applied optimized surrogate. explore time-dependent propagation outputs. key identified calculating Sobol indices accurate efficient could be readily implemented field-scale