Modeling and forecasting subsurface multiphase fluid flow fields underpin applications ranging from geological CO2 sequestration (GCS) operations to geothermal production. This is essential for ensuring both operational performance and long-term safety. While high fidelity multiphase simulators are widely used for this purpose, they become prohibitively expensive once many forward runs are required for inversion purposes and quantify uncertainty. To tackle this challenge we propose LAViG-FLOW, a latent autoregressive video generation diffusion framework that explicitly learns the coupled evolution of saturation and pressure fields. Each state variable is compressed by a dedicated 2D autoencoder, and a Video Diffusion Transformer (VDiT) models their coupled distribution across time. We first train the model on a given time horizon to learn their coupled relationship and then fine-tune it autoregressively so it can extrapolate beyond the observed time window. Evaluated on an open-source CO2 sequestration dataset, LAViG-FLOW generates saturation and pressure fields that stay consistent across time while running orders of magnitude faster than traditional numerical solvers.

翻译：对地下多相流体流动场进行建模与预测，是支撑从地质二氧化碳封存（GCS）作业到地热生产等一系列应用的基础。这对于确保操作性能与长期安全至关重要。虽然高保真度多相模拟器被广泛用于此目的，但一旦需要大量正演模拟以用于反演目的和量化不确定性时，其计算成本将变得极其高昂。为应对这一挑战，我们提出了LAViG-FLOW，这是一个潜在自回归视频生成扩散框架，它显式地学习饱和度场与压力场的耦合演化过程。每个状态变量通过一个专用的二维自编码器进行压缩，并由一个视频扩散Transformer（VDiT）对它们随时间变化的耦合分布进行建模。我们首先在给定的时间范围内训练模型以学习其耦合关系，然后以自回归方式对其进行微调，使其能够外推到观测时间窗口之外。在一个开源CO2封存数据集上的评估表明，LAViG-FLOW生成的饱和度场和压力场在时间上保持一致，同时运行速度比传统数值求解器快数个数量级。