Fault diagnosis algorithm of operating mechanism based on Bayesian optimized support vector machine

Mechanical failure is one of the main faults in High Voltage Circuit Breakers (HVCB), and the mechanical characteristic parameters of HVCB directly determine their mechanical state. In this article, to efficiently and accurately identify the mechanical state of circuit breakers, we use a Bayesian optimization Support Vector Machine (SVM) algorithm to diagnose the mechanical characteristics of the circuit breaker. By analyzing the features of the opening and closing coils and the contact stroke, features are extracted according to the timing characteristics to construct a dataset. Based on the principle of the Bayesian optimization algorithm, the SVM is optimized to obtain the optimal combination of parameters for the SVM model’s performance. Experimental results show that under the Bayesian optimization algorithm, the optimal parameter combination enables the SVM model to achieve an accuracy of 0.988. This method provides a new approach for the diagnosis and fault warning of HVCB.