基于核主成分分析-SoftMax的高压断路器机械故障诊断技术研究

The high voltage circuit breaker (HVCB) is a crucial equipment to ensure the security and reliability of power system, consequently the mechanical fault diagnosis research of HVCB has become a key issue. In this paper, a SoftMax classifier model based on Kernel principal component analysis (KPCA) was developed, which was located to identify the vibration signal of typical working conditions. Firstly, the wavelet packet time-frequency energy rate was adopted as the characteristic description of six typical mechanical conditions. Secondly, KPCA was used for dimensionality reduction to obtain a feature space with lower latitude and high-recognition. Then, SoftMax was adopted to diagnose the typical working conditions. To prove the superiority of the SoftMax diagnostic model combined with KPCA feature space, the comparative experiment of SoftMax classifier results in the origin feature space, the principal component analysis (PCA) feature space, the KPCA feature space was carried out, the comparison for accuracy of various methods in the KPCA feature space was proceed as assist. The result indicates that the proposed method provides a new thought for HVCB mechanical fault diagnosis.