Rayleigh-Bénard Convection Control by Reinforcement Learning

As a self-driven project, I developed a reinforcement learning framework that controls Rayleigh–Bénard convection by adjusting the bottom wall temperature over time. The idea comes from the study by Beintema et al. To implement it, I coupled OpenFOAM with Python and used the Proximal Policy Optimization (PPO) algorithm to learn control strategies for the flow.