Neural networks for power flow: Graph neural solver

• Published in: Electric power systems research Vol. 189; p. 106547
• Main Authors: Donon, Balthazar, Clément, Rémy, Donnot, Benjamin, Marot, Antoine, Guyon, Isabelle, Schoenauer, Marc
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.12.2020; Elsevier Science Ltd; Elsevier
• Subjects: Alternating current; Artificial Intelligence; Artificial neural networks; Computer architecture; Computer Science; Computing time; Electric power grids; Graph neural networks; Graph neural solver; Graphs; Neural networks; Power flow; Probability distribution; Solver; Solvers; Topology; Solver; Graph neural solver; Graph neural networks; Power flow; Artificial neural networks; Power Flow; Artificial Neural Networks; Graph Neural Solver; Graph Neural Networks
• ISSN: 0378-7796; 1873-2046
• DOI: 10.1016/j.epsr.2020.106547

•Graph neural networks to solve AC power flow.•Our graph neural solver uses a novel graph neural network architecture.•Training is performed by minimizing the violation of physical laws.•It is robust to variations of injections, power grid topology, and line characteristics.•Experimental validation on standard IEEE power grids (case9, case14, case30, case118). Recent trends in power systems and those envisioned for the next few decades push Transmission System Operators to develop probabilistic approaches to risk estimation. However, current methods to solve AC power flows are too slow to fully attain this objective. Thus we propose a novel artificial neural network architecture that achieves a more suitable balance between computational speed and accuracy in this context. Improving on our previous work on Graph Neural Solver for Power System [1], our architecture is based on Graph Neural Networks and allows for fast and parallel computations. It learns to perform a power flow computation by directly minimizing the violation of Kirchhoff’s law at each bus during training. Unlike previous approaches, our graph neural solver learns by itself and does not try to imitate the output of a Newton-Raphson solver. It is robust to variations of injections, power grid topology, and line characteristics. We experimentally demonstrate the viability of our approach on standard IEEE power grids (case9, case14, case30 and case118) both in terms of accuracy and computational time.