Diagnosis Method for Commutation Failure Based on Gramian Angular Field and PCNN-GRU
High voltage direct current transmission as an efficient power transmission technology,the commutation failure in its operation will lead to a rapid increase in DC current and a sharp drop in DC voltage,which have a significant impact on the safe and stable operation of the power grid.For commutation failure,a diagnosis method for commutation failure which combined Gramian angular field(GAF)with parallel convolutional neural network-gated recurrent unit(PCNN-GRU)is proposed.The one-dimensional time series signal is converted into two-dimensional image feature map using GAF,which preserves the time series information of the signal.The feature extraction capability of convolutional neural network and the time series feature processing capability of gated recurrent unit of PCNN-GRU model are then used to enable the model to learn more fault features and improve the diagnostic performance of the model.The experiment on Yongfu DC transmission system shows that the diagnostic accuracy of this method is 99.33%,with strong multi-feature extraction ability and time series feature analysis ability,and the diagnostic performance is strong,which can quickly respond to and identify commutation failure.
high voltage direct current transmissioncommutation failureGramian angular fieldPCNN-GRUfault diagnosisdeep learning
万方数据
维普
