Power Flow Feasibility Assessment Using Variational Graph Autoencoders

2026-07-10

AI Summary

This study addresses the challenge of effectively evaluating the physical feasibility of AI-generated solutions in data-driven power flow analysis. To this end, it introduces, for the first time, a variational graph autoencoder (VGAE) for power flow feasibility assessment, leveraging graph neural networks to explicitly model the complex relationships between power grid topology and operational states. Experimental results on the IEEE 118-bus system demonstrate that the proposed method achieves high accuracy in identifying infeasible solutions, thereby significantly enhancing the reliability of data-driven power flow solvers. This work fills a critical research gap in solution feasibility verification within the domain of AI-based power system analysis.

Abstract

Data-driven methods, including graph neural networks, have been studied for accelerating power flow calculations in recent years, but very little attention has been paid to the solution feasibility, which can be obtained by traditional solvers. This paper presents a Variational Graph Autoencoder (VGAE) that detects the power flow solution feasibility, using the IEEE 118-bus case, to assess the validity of the solutions provided by AI-driven solvers.

Problem

• Power Flow Feasibility
• Graph Neural Networks
• AI-driven Solvers
• Variational Graph Autoencoder

Innovation

• Variational Graph Autoencoder
• Power Flow Feasibility
• Graph Neural Networks
• Data-driven Power Systems
• Solution Validity