Remaining Life Prediction of Conventional Low-Voltage Circuit Breaker Contact System Based on Effective Vibration Signal Segment Detection and MCCAE-LSTM

In order to realize on-line mechanical life prediction of conventional low-voltage circuit breaker and improve its operation reliability, a life prediction method based on vibration signal detection is proposed. Firstly, the effective vibration segment that characterizes the mechanical performance of contact system are extracted by combining with the designed signal processing method; then, a multi-channel convolutional auto-encoder long short term memory network (MCCAE-LSTM) life prediction model for individual circuit breaker is constructed, in which the MCCAE is used to obtain the time sequential micro features rich in life degradation information, and the LSTM realizes the quantitative output of the remaining mechanical life; finally, through the actual mechanical life experiment and prediction performance analysis, the effective vibration segment can reflect the mechanical characteristics of the contact system, the measurement method is easy to be carried out on-line, and the average goodness of fit of the prediction model is improved by about 10% compared with the traditional method, which verify the effectiveness of the proposed method.