Virtual: https://events.vtools.ieee.org/m/538226

The Optimal Power Flow (OPF) problem is one of the most critical and computationally intensive optimization challenges in power systems, often requiring solutions within very short time intervals, such as five-minute market windows. To address this challenge, we first employ machine learning techniques to approximate the nonlinear power balance equations with linear models, significantly reducing computational complexity. In parallel, we reformulate the OPF problem using shortest-path methods, transforming it into a graph-based optimization framework. This approach enables the computation of high-quality solutions without relying on commercial solvers, while achieving accuracy comparable to state-of-the-art solvers such as <a href="http://Gurobi.Co-sponsored" target="_blank" title="Gurobi.Co-sponsored">Gurobi.Co-sponsored by: Resilience and Clean Energy Systems (RCES)Speaker(s): Sajad Fathi HafshejaniVirtual: https://events.vtools.ieee.org/m/538226