GIS Mechanical Fault Diagnosis and Test Analysis Based on S Transform-SVM-AlexNet Model

To effectively test Gas Insulated Switchgear (GIS) mechanical faults, test simulation systems of three kinds of that were installed. Vibration signals were measured by the opening and closing operation of the GIS circuit breaker as an excitation source and a GIS mechanical fault diagnosis method based on S Transform-SVM-AlexNet Model was proposed. Vibration signals were processed by S Transform and the time-frequency diagram containing equipment characteristics was obtained. The S Transform-SVM-AlexNet Model was established and the pre-trained AlexNet Neural Network Model was used to extract S Transform image features as predictive variables. Support Vector Machine (SVM) was fitted for image pre-classification. According to the classification result of the fuzzy matrix, effective measuring points were screened out. Time-frequency diagrams of effective measuring points were sent into AlexNet Model for transfer learning, the fine-tuned neural network model obtained, which can realize the effective diagnosis of GIS mechanical faults.