Partial Discharge Detection and Fault Identification of Switchgear Based on Improved Deep Learning
Aiming at the problem that traditional switchgear partial discharge statistical feature extraction needs to rely on ex-pert experience and lack generalization ability,a switchgear partial discharge detection and fault identification model based on improved DBN-LSTM deep learning is proposed.In this model,the deep belief network(DBN)can directly and autonomously extract the global effective feature information of the sample and the long short-term memory network(LSTM)is good at min-ing the time domain information of the statistical feature and the Dropout technology is used to reduce the influence of DBN overfitting.It improves the generalization ability of the model.Combined with the defect spectrum of four typical switchgear partial discharge models,the performance of the proposed model is tested,and compared with other algorithms,the results show that the proposed algorithm has a good effect on the fault identification of switchgear,and the comprehensive fault accura-cy rate reaches 97％,the overall recognition performance of the proposed model is better than the single DBN and LSTM and DBN-LSTM models.
deep learningswitchgearpartial discharge detectionDBN-LSTMfault identification
万方数据
维普
