Mechanical Fault Diagnosis in High Voltage Vacuum Circuit Breaker Based on Improved S Transform and Support Vector Machine

In recent years, a new electromagnetic repulsion mechanism (ERM) has been applied in high voltage vacuum circuit breakers (HVVCBs). However, the current research on ERM mainly focuses on the design and improvement of mechanical structures, ignoring the aspect of their fault diagnosis. In order to determine the fault types occurred in ERM, a fault diagnosis method based on improved S transform (IST) and support vector machine (SVM) is proposed. Firstly, the vibration signals are obtained at two different positions of the HVVCB with ERM during the operating process. Then, the IST is used to conduct the time-frequency analysis on the vibration signals. The feature is extracted based on the energy entropy from the normalized energy. Finally, the grid search (GS) and particle swarm optimization (PSO) algorithms are adopted to realize parameters optimization of support vector machine (SVM). Moreover, the other two features and three classifiers are used to verify the effectiveness of IST. The results show that the proposed method is suitable for HVVCB mechanical fault diagnosis.