Data-Driven Fault Parameter Extraction and Fault Diagnosis Model for High-Voltage Circuit Breakers

With the development of the new power system, the integration of information and physics has gradually deepened, effectively solving the problems of condition monitoring and fault diagnosis in all aspects of the power system. High voltage circuit breakers are a vital component of the new power system, and once a fault occurs, it may seriously affect the safe and stable operation of the power system. To solve the current problem of inaccurate mechanical fault diagnosis of high-voltage circuit breakers, this paper proposes a data-driven method for extracting fault feature parameters of high-voltage circuit breakers and establishes a fault diagnosis model based on an improved SVDD (Support Vector Data Description) algorithm. This study uses wavelet packet decomposition to extract two eigenvectors, wavelet node entropy and relative energy value, from high-voltage circuit breaker vibration signals, and also proposes a dimensionality reduction algorithm based on kernel function entropy component analysis, which effectively improves the efficiency of eigenparameter extraction. Then, an improved SVDD algorithm is proposed for fast and accurate classification and identification of high-voltage circuit breaker fault types. Finally, the simulation algorithm is validated based on the actual fault data set of high-voltage circuit breakers. Through experiments, we have confirmed that the method proposed in this paper achieves significant results in identifying different types of mechanical faults in high-voltage circuit breakers, which is essential to ensure the safe and stable operation of power systems.