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

## TL;DR

This paper introduces a machine learning method that uses a differentiable multiphase flow simulator to efficiently manage reservoir pressure in subsurface systems.

## Contribution

A novel physics-informed machine learning workflow that reduces the computational cost of reservoir pressure predictions using transfer learning.

## Key findings

- The model achieves high accuracy with fewer than three thousand full-physics simulations, a drastic reduction from previous estimates.
- Transfer learning from single-phase simulations significantly speeds up training for complex multiphase scenarios.
- The CNN effectively predicts fluid extraction rates to maintain pressure limits in heterogeneous reservoirs.

## 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 compared to previous works. To speed up training, we pretrain the model on single-phase, steady-state simulations and then finetune 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.