Transient Voltage Stability Assessment Method for Power System Data Missing
Aiming at the problem of decreased accuracy of transient voltage stability assessment model when data is missing,the author proposes a transient voltage stability assessment method for power system based on multi-view missing data filling and gating graph neural network.Firstly,the missing data are filled based on the complementary spatio-temporal views of multiple views to obtain a complete dataset.Then,the restored complete dataset is used to train the gating graph neural network model for transient voltage stability assessment,and the assessment model should be updated quickly to improve the performance of online applications.Finally,the effectiveness of the proposed method is verified on the IEEE39-node system example.The simulation results show that the proposed method can fill the missing data in a timely and effective manner in case of any synchronization vector measurement unit placement information loss and network topology changes,and the evaluation performance of the used evaluation model has significant advantages.
measurement data missingspatio-temporal viewgating graph neural networktransient voltage stability assessment
万方数据
维普
