Oil extraction requires an understanding of underground pressure and the distribution of oil and water at depths of several kilometers. However, due to rock heterogeneity and multiphase fluid interference, accurate prediction is extremely challenging.

According to a study published in Expert Systems with Applications, a research team from the Shenyang Institute of Automation of the Chinese Academy of Sciences has proposed a novel artificial intelligence framework, termed PI-DeepOKAN. This framework deeply integrates physical laws and provides an efficient new tool for reservoir dynamic simulation and intelligent decision-making.

Traditional purely data-driven models rely heavily on large amounts of labeled samples and struggle to guarantee physical consistency and generalization capability, while conventional numerical simulation methods are computationally time-consuming and cannot meet the engineering demands for real-time prediction and rapid optimization.

The researchers constructed a dual-branch deep operator architecture with output-head reweighting to achieve a unified representation of static geological parameters and dynamic production conditions. Recognizing the distinct evolution patterns of the pressure and saturation fields, the researchers introduced an attention mechanism that adaptively adjusts the fusion weights of geological parameters and well-control information, overcoming the limitation of fixed multi-source information fusion found in conventional deep operator networks.

The framework also integrates the nonlinear mapping capability of the Kolmogorov–Arnold Network (KAN), significantly enhancing the model's ability to characterize complex multiphase flow behavior in highly heterogeneous reservoirs under limited sample conditions.

In addition, the researchers constructed a full-chain physics-informed loss function that encompasses multiphase flow governing equations, initial conditions, internal and external boundary conditions, and fault interface constraints, ensuring that the model predictions strictly conform to reservoir flow mechanisms.

Experimental results demonstrate that PI-DeepOKAN can accurately capture pressure propagation and waterflood front evolution in complex reservoirs. While maintaining physical consistency, it dramatically improves prediction efficiency, serving as a highly efficient surrogate model for rapid reservoir simulation, production performance analysis, and intelligent optimization decision-making.