RT25 | Accelerating Reinforcement Learning Training for Power Systems Applications Using OPAL-RT

Dr. Timothy M. Hansen from Colorado State University presents a real-time reinforcement learning (RL) framework for grid frequency control using OPAL-RT simulation. Demonstrated on Alaska’s Cordova microgrid, the approach enables 20× faster RL training, improving fast frequency response (FFR) with BESS integration. The study bridges AI and power systems through scalable, real-time co-simulation.