Speaker
Description
The Lattice Boltzmann Method (LBM) has evolved into one of the most versatile computational frameworks for simulating complex fluids and multiscale flow phenomena. Its kinetic formulation naturally accommodates multiphase and multicomponent flows, thermal transport, fluid–structure interactions, porous media flows, and turbulence, while offering excellent scalability on modern high-performance computing architectures.
This talk provides an overview of the current state of the art in advanced Lattice Boltzmann methods, with a particular focus on recent developments in collision operators and higher-order kinetic models.
Moreover, we will cover opportunities at the intersection between kinetic theory and machine learning. We will discuss the potential of learning novel collision operators using neural networks constrained by fundamental physical principles, including conservation laws and symmetry requirements. This physics-informed approach may open the way to a new generation of Lattice Boltzmann methods with enhanced stability and accuracy for complex fluid and multiscale flow simulations.