An online power system transient stability assessment method based on graph neural network and central moment discrepancy

The increasing penetration of renewable energy introduces more uncertainties and creates fluctuations in power systems. Conventional offline time-domain simulation-based stability assessment methods may no longer be able to face changing operating conditions. In this work, a graph neural network-based online transient framework is proposed, which can interactively work with conventional provide results. proposed consists feature preprocessing module, multiple physics-informed networks, an updating scheme transfer learning central moment discrepancy. t-distributed stochastic neighbor embedding used virtualize the effectiveness framework. IEEE 16-machine 68-bus system for case studies. results show that method achieve accurate under conditions