Application of Kernel Principal Components Analysis to Fault Diagnosis of High-Voltage Circuit Breakers

In order to meet the development trend of smart grids and improve fault diagnosis precision of high-voltage circuit breakers(HVCBs),kernel principal components analysis(KPCA) is applied in fault diagnosis of HVCBs.As an advanced intelligent algorithm,KPCA has great advantages in solving nonlinear problems.Once the proper kernel function is selected,KPCA can realize the mapping of the original data to the feature space.Afterwards,fault diagnosis can be accomplished in the feature space through principal component calculation and square prediction error(SPE) statistics.To test the effectiveness of the proposed approach,normal and anormal data collected by an on-line monitoring system are analyzed by the KPCA program and Matlab simulation,through which fault data can be diagnosed.The results have proved that KPCA method can obtain a better diagnostic precision with high stability.