Abstract:To maintain a reliable grid we need fast decision-making algorithms for complex problems like Dynamic Reconfiguration (DyR). DyR optimizes distribution grid switch settings in real-time to minimize grid losses and dispatches resources to supply loads with available generation. DyR is a mixed-integer problem and can be computationally intractable to solve for large grids and at fast timescales. We propose GraPhyR, a Physics-Informed Graph Neural Network (GNNs) framework tailored for DyR. We incorporate essential operational and connectivity constraints directly within the GNN framework and train it end-to-end. Our results show that GraPhyR is able to learn to optimize the DyR task.
Abstract:A transformation of the US electricity sector is underway with aggressive targets to achieve 100% carbon pollution-free electricity by 2035. To achieve this objective while maintaining a safe and reliable power grid, new operating paradigms are needed, of computationally fast and accurate decision making in a dynamic and uncertain environment. We propose a novel physics-informed machine learning framework for the decision of dynamic grid reconfiguration (PhML-DyR), a key task in power systems. Dynamic reconfiguration (DyR) is a process by which switch-states are dynamically set so as to lead to an optimal grid topology that minimizes line losses. To address the underlying computational complexities of NP-hardness due to the mixed nature of the decision variables, we propose the use of physics-informed ML (PhML) which integrates both operating constraints and topological and connectivity constraints into a neural network framework. Our PhML approach learns to simultaneously optimize grid topology and generator dispatch to meet loads, increase efficiency, and remain within safe operating limits. We demonstrate the effectiveness of PhML-DyR on a canonical grid, showing a reduction in electricity loss by 23%, and improved voltage profiles. We also show a reduction in constraint violations by an order of magnitude as well as in training time using PhML-DyR.