Advances in data-driven methods for Coupled Fluid Flow and Transport

Alvaro Coutinho

Universidade Federal do Rio de Janeiro

In recent years, there has been significant interest in using data-driven methods to solve problems in science and engineering, especially in the context of large coupled fluid flow and transport. Numerical simulations for these problems can be costly, making data-driven methods valuable for understanding and improving efficiency in quantifying and predicting states. This talk will review recent advancements in data-driven methods, such as dynamic mode decomposition, physics-informed neural networks, manifold learning, and neural operators, as applied to relevant problems involving coupled incompressible fluid flow with transport. These problems are of interest in sustainable resource exploration, geophysics, and various industrial applications. The talk will show how data-driven information can improve the efficiency of numerical simulation software for short-time prediction and adaptive time-stepping strategies, exploring parametric manifolds for unseen scenarios, and reconstructing high-dimensional simulations with lower-dimensional structures in feasible time.