Bearing fault diagnosis method based on ngas synchronously optimizing wavelet filter and mckd

Abstract

The invention discloses a bearing fault diagnosis method for NGAs synchronously optimizing wavelet filters and MCKD. First, use the sensor to collect the original vibration signal and input it, set the initial conditions of the Niche Genetic Algorithm (NGAs), and use the NGAs to carry out the Morlet wavelet filter center frequency and bandwidth, maximum correlation kurtosis deconvolution (MCKD) filter length and period Synchronous joint optimization, taking the correlation kurtosis (CK) of the occurrence characteristics of the bearing fault shock cycle as the optimization index, realizes the parameter adaptive synchronization optimization of the two processing steps before and after, adopts Morlet band-pass filter preprocessing, and MCKD performs in-band filtering on the filtered signal Noise reduction processing, and finally use the envelope spectrum of the MCKD in-band noise reduction signal to determine whether there is a fault and the type of fault. The analysis of simulation signal, laboratory signal and experimental data of Dongfang Institute shows that the method proposed in this paper can effectively eliminate the influence of external occasional interference impact and reduce the influence of signal transmission path and noise, ensuring the effectiveness of bearing fault diagnosis.

Description

technical field [0001] The invention relates to a NGAs synchronously optimized wavelet filter and MCKD bearing fault diagnosis method, which belongs to the technical field of rolling bearing fault diagnosis. Background technique [0002] As one of the most important components of rotating machinery, rolling bearings are widely used in important fields such as machinery, transportation, aerospace, etc., and are also the most prone to failure due to harsh working conditions. Once the rolling bearing fails and is not found in time, it may cause immeasurable consequences. Therefore, how to accurately judge the health status of rolling bearings is very important to improve the reliability and availability of mechanical equipment and ensure the safe operation of equipment. However, vibration signals are often submerged in strong background noise and occasional high-amplitude interference shocks, making it difficult to extract fault feature information. Therefore, the key to accu...

AI Technical Summary

Problems solved by technology

[0003] The optimization index adopted in the processing of band-pass filtering and in-band noise elimination does not take into account the periodic occurrence characteristics of bearing transient fault shocks, which makes it susceptible to occasional interference shocks and the two processing steps are independently optimized, which is difficult to guarantee the accuracy of diagnosis The problem of the overall effect, the present invention proposes a kind of Morlet wavelet filtering and the maximum correlation kurtosis deconvolution (MCKD) parameter synchronous optimization bearing fault diagnosis method

Images