Running Status Diagnosis of Onboard Traction Transformers Based on Kernel Principal Component Analysis and Fuzzy Clustering

The onboard traction transformer is a critical equipment of high-speed trains, its running state directly affects the safety and stability train’s operation. Given complexity condition transformer, this paper proposes diagnosis algorithm based on kernel principal component analysis (KPCA) fuzzy clustering. To fully extract status information aging characteristics insulating oil main insulation are analyzed under different mileage as first step. Thereby, to eliminate signal redundancy, feature set by KPCA combined with characteristic quantities traditional dissolved gas (DGA), eigenvalues contribution rate over 95% used new eigenvectors. Finally, model established using clustering analysis, considering limitations fault data transformer. results from field collected show that proposed method effective in diagnosing