Deep learning-based anomaly detection and fault prediction method for permanent magnet fast ring network cabinet

The ring network cabinet of the distribution network is an important part of the urban power system, and its operation state directly affects the stability and reliability of the power system.In this paper, a deep learning algorithm is used to analyze and process the partial discharge signal, and a permanent magnet fast ring main unit partial discharge detection and fault identification model based on improved DBN-LSTM is proposed.By analyzing a large amount of local discharge signal data under normal operation and fault conditions of ring main cabinet, and using these data to train a deep learning-based fault prediction model.The performance of the improved DBN-LSTM model is tested by combining the defect spectrograms of four typical ring network cabinet partial discharge models and compared with other algorithms.The proposed model has good effect on fault identification of ring network cabinet, with a combined identification accuracy of 98.41%, and the overall identification performance is better than both BP neural networks and SVM classifiers.The prediction accuracy of the fault prediction model also reaches 88.52%, and the experimental results of the method in this paper are more satisfactory.