Mechanical fault diagnosis of on-load tap-changer based on improved variational mode decomposition and relative density-based outlier score

In order to diagnose the mechanical fault of transformer on-load tap-changer (OLTC) effectively, a fault diagnosis method based on improved variational mode decomposition (VMD) and relative density-based outlier score (RDOS) is proposed. Firstly, the signal is decomposed into a series of modal components with narrow band and distinguishing centre frequency by improved VMD based on optimal quality factor. Then, the energy, permutation entropy, power spectrum concentration degree and characteristic frequency of each modal component are calculated, and Laplacian-score method is used to select the features with better discriminative ability. Finally, a RDOS is used to diagnose the mechanical fault of OLTC, and the mechanical condition is judged according to RDOS value. An experiment is conducted on the OLTC simulation experiment platform and the collected signals are processed. The result shows that the proposed method can diagnose mechanical fault effectively.