Identification and correction of abnormal measurement data in power system based on graph convolutional network and gated recurrent unit

•A GCN fusion GRU method for extracting spatiotemporal features of measurement data is proposed, which can maximize the recovery of the true features of abnormal data. Compared with existing methods, it has higher data reconstruction accuracy. The maximum relative RMSE of 4-dimensional state quantity does not exceed over 0.006.•Pre-identification process based on interval slope is proposed to locate the fuzzy intervals of the predicted data, which can greatly reduce the overall running time of the identification process. The test results show that the running time after introducing the pre-identification process has been reduced from 3.55 s to 1.89 s.•The effectiveness and superiority of the proposed method were verified through simulation experiments and actual data testing. The experimental results show that the proposed method has better performance and higher accuracy than existing methods.