The high frequency vibration signal in the operating mechanism of the circuit breaker contains rich me-chanical information. In order to effectively extract the high frequency component of the vibration for diagnosis andanalysis

a new mechanical fault diagnosis method of the operating mechanism of the circuit breaker is proposed.This method decomposes the collected vibration signal in time domain through CEEMDAN

and reconstructs its highfrequency components according to entropy weight method to get the denoised signal

decomposes the denoised sig-nal with LMD

and calculates the multiscale permutation entropy of the decomposed pf component as the input char-acteristic component of KFCM recognition algorithm. The results of KFCM identification show that this method has ahigh identification rate for typical faults such as shaft jam

base looseness

and misoperation.