AI as a Technology Enabler for Power Systems Applications

Pascal Van Hentenryck – Georgia Tech and Gurobi Optimization

In many industry power systems applications, the same optimization problem is solved repeatedly for instances taken from a distribution that can be learned or forecasted. This talk studies how to speed up these parametric optimization problems to meet real-time constraints present in many applications. It reviews the concept of optimization proxies that learn the input/output mappings of parametric optimization problems. This novel methodology is demonstrated on industrial problems in grid optimization. These results show that the fusion of AI and Optimization can deliver outcomes that the two technologies cannot achieve independently.