A Graph Neural Network Framework for Grid-Based Simulation

TL;DR

This paper introduces a graph neural network framework that acts as a surrogate model to replace expensive reservoir simulations, significantly speeding up well placement optimization in subsurface applications.

Contribution

The paper presents a novel GNN-based surrogate model for reservoir simulation, enabling faster optimization in well control and placement tasks.

Findings

GNN model achieves close match with simulation results in predictions.

Model trained on 6000 samples, tested on 6000 samples.

Potential applications in oil, gas, and carbon sequestration.

Abstract

Reservoir simulations are computationally expensive in the well control and well placement optimization. Generally, numerous simulation runs (realizations) are needed in order to achieve the optimal well locations. In this paper, we propose a graph neural network (GNN) framework to build a surrogate feed-forward model which replaces simulation runs to accelerate the optimization process. Our GNN framework includes an encoder, a process, and a decoder which takes input from the processed graph data designed and generated from the simulation raw data. We train the GNN model with 6000 samples (equivalent to 40 well configurations) with each containing the previous step state variable and the next step state variable. We test the GNN model with another 6000 samples and after model tuning, both one-step prediction and rollout prediction achieve a close match with the simulation results. Our GNN framework shows great potential in the application of well-related subsurface optimization including oil and gas as well as carbon capture sequestration (CCS).