Differentiable multiphase flow model for physics-informed machine learning in reservoir pressure management

TL;DR

This paper presents a physics-informed machine learning approach combining a differentiable multiphase flow simulator with a CNN to predict reservoir pressure management efficiently, significantly reducing the need for costly simulations.

Contribution

It introduces a novel workflow coupling a differentiable multiphase flow simulator with CNNs, utilizing transfer learning to drastically cut down simulation requirements for accurate reservoir predictions.

Findings

Achieves high-accuracy predictions with fewer than 3,000 simulations.

Reduces computational cost compared to previous methods requiring up to ten million simulations.

Demonstrates effective transfer learning from single-phase to multiphase flow scenarios.

Abstract

Accurate subsurface reservoir pressure control is extremely challenging due to geological heterogeneity and multiphase fluid-flow dynamics. Predicting behavior in this setting relies on high-fidelity physics-based simulations that are computationally expensive. Yet, the uncertain, heterogeneous properties that control these flows make it necessary to perform many of these expensive simulations, which is often prohibitive. To address these challenges, we introduce a physics-informed machine learning workflow that couples a fully differentiable multiphase flow simulator, which is implemented in the DPFEHM framework with a convolutional neural network (CNN). The CNN learns to predict fluid extraction rates from heterogeneous permeability fields to enforce pressure limits at critical reservoir locations. By incorporating transient multiphase flow physics into the training process, our method enables more practical and accurate predictions for realistic injection-extraction scenarios compare to previous works. To speed up training, we pretrain the model on single-phase, steady-state simulations and then fine-tune it on full multiphase scenarios, which dramatically reduces the computational cost. We demonstrate that high-accuracy training can be achieved with fewer than three thousand full-physics multiphase flow simulations -- compared to previous estimates requiring up to ten million. This drastic reduction in the number of simulations is achieved by leveraging transfer learning from much less expensive single-phase simulations.