In order to improve the accuracy of fault diagnosis for the spring operating mechanism of high-voltage circuit breakers

a fault diagnosis method based on the joint analysis of multi-source signal temporal characteristics

such as closing and opening coil currents

travel

and vibration

is proposed. This paper takes the CT-26 circuit breaker as the research object. Through equivalent simulation

the sequence relationship of component movements is obtained

the component movement states are acquired

and combined with the signal temporal characteristics

and the signals are divided into five temporal segments including the energized stage

the action tripping stage

and the acceleration movement stage. The fault signals are segmented according to the simulation temporal relationship. Different entropy information features of vibration signals in different temporal windows are obtained

as well as the time-domain and energy features of travel signals and current signals. Secondly

by fusing the single signal characteristics of different temporal windows

the support vector machine model (support vector machine

SVM) is used to identify the abnormal state of the single signal

and the Newton-Raphson-based optimizer (NRBO) is used to optimize the model parameters. Finally

the posterior probability of SVM is weighted

the Dempster-Shafer (D-S) evidence theory is used to fuse the multi-source signal samples

and the final diagnosis result is calculated. The results of the case analysis show that the recognition algorithm based on the temporal characteristics of multi-source signals and NRBO-SVM-DS has an accuracy rate of 97.5%

effectively improving the diagnostic accuracy.