Research on Adaptive Fault Diagnosis Method for Multi-type Circuit Breakers

Abstract

The opening/closing coil current waveform of circuit breaker varies with the manufacturers and the voltage levels and shows different morphological characteristics. The traditional circuit breaker fault identification is mainly based on the key extreme points of the coil current waveform, however, some extremum points in the actual waveform may be missing, which can easily lead to inaccurate extraction of the key points, or even misjudgment. To solve the problem, firstly, this paper proposes a method to extract the key points based on sliding window detection, the method can adaptively and accurately locate the key points for different current waveforms; Secondly, a fault diagnosis model that integrates principal component analysis (PCA)、grid search(GS) and support vector machine (SVM) is proposed. Finally, the key point positioning algorithm performance of sliding window method and the fault identification effects based on PCAGS-SVM are experimentally verified. The sliding window detection method is suitable for the extreme point positioning of the opening/closing current of different kinds of circuit breakers. Furthermore, compared with the traditional identification models, the fault diagnosis model proposed in this paper has a fault classification accuracy rate of 96.67% and 95.00% in the case of opening/closing statues, respectively, fault identification accuracy is improved.