At present, in the fault diagnosis method of on-load tap-changer, it is unreasonable to directly characterize the on-load tap-changer as normal or fault, ignoring that the occurrence of a fault is a gradual process. In this paper, three kinds of electrodes are designed to simulate the fault discharge development process of the on-load tap-changer. Based on the chromatographic data of oil sample detection, by using the TOPSIS method, the state of the on-load tap-changer is set to negative ideal solution and positive ideal solution. Calculate the degree of closeness, using the oil chromatographic data as the reference sequence, and calculate of grey correlation degree. The degree of failure is classified, the larger the quantified value of the failure state, the greater the possibility of failure. This method can reduce the equipment failure rate.

TOPSIS;grey correlation relativity;discharge experiments;fault degree classification