Deep Learning-Accelerated Surrogate Optimization for High-Dimensional Well Control in Stress-Sensitive Reservoirs

2026-03-31

AI Summary

This study addresses the challenge of optimizing high-dimensional, time-varying well control strategies in stress-sensitive unconventional reservoirs, where nonlinear coupling between pressure-driven flow and stress-induced permeability degradation renders conventional optimization methods computationally prohibitive due to their reliance on expensive fully coupled flow–geomechanics simulations. To overcome this limitation, the authors propose a deep learning surrogate model that maps bottomhole pressure trajectories to cumulative production via a neural network embedded within a constrained optimization framework. A novel problem-oriented sampling strategy is introduced to generate training data aligned with optimal control trajectories, thereby enabling synergistic alignment between the surrogate model and the optimization process. Numerical experiments demonstrate that the method achieves prediction errors of only 2–5% across multiple initializations while reducing computational cost by three orders of magnitude, substantially enhancing the scalability and reliability of PDE-constrained optimization in complex reservoir systems.

Technology Category

Search and Optimization: Sampling/Simulation-based SearchReasoning under Uncertainty: Stochastic OptimizationConstraint Satisfaction and Optimization: Constraint Learning and Acquisition

Application Category

Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingSystems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Production optimization in stress-sensitive unconventional reservoirs is governed by a nonlinear trade-off between pressure-driven flow and stress-induced degradation of fracture conductivity and matrix permeability. While higher drawdown improves short-term production, it accelerates permeability loss and reduces long-term recovery. Identifying optimal, time-varying control strategies requires repeated evaluations of fully coupled flow-geomechanics simulators, making conventional optimization computationally expensive. We propose a deep learning-based surrogate optimization framework for high-dimensional well control. Unlike prior approaches that rely on predefined control parameterizations or generic sampling, our method treats well control as a continuous, high-dimensional problem and introduces a problem-informed sampling strategy that aligns training data with trajectories encountered during optimization. A neural network proxy is trained to approximate the mapping between bottomhole pressure trajectories and cumulative production using data from a coupled flow-geomechanics model. The proxy is embedded within a constrained optimization workflow, enabling rapid evaluation of control strategies. Across multiple initializations, the surrogate achieves agreement with full-physics solutions within 2-5 percent, while reducing computational cost by up to three orders of magnitude. Discrepancies are mainly associated with trajectories near the boundary of the training distribution and local optimization effects. This framework shows that combining surrogate modeling with problem-informed sampling enables scalable and reliable optimization for high-dimensional, simulator-based problems, with broader applicability to PDE-constrained systems.

Problem

• stress-sensitive reservoirs
• production optimization
• flow-geomechanics coupling
• high-dimensional control
• computational cost

Innovation

• surrogate modeling
• deep learning
• problem-informed sampling
• high-dimensional optimization
• flow-geomechanics coupling