Vibration signal identification method and system based on improved multi-scale permutation entropy

Abstract

The invention provides a vibration signal identification method and system based on improved multi-scale permutation entropy. The method comprises the following steps: collecting vibration signals in different working states, and carrying out multi-scale analysis based on a sliding window and a sample quantile on the collected vibration signals in different working states to obtain a sequence after the multi-scale analysis; solving the permutation entropy of the sequence after the multi-scale analysis to obtain a multi-scale permutation entropy value, and taking the multi-scale permutation entropy value as the feature vector of the vibration signal in different fault states; and inputting the feature vector into a classifier based on machine learning, thereby realizing identification of vibration signals in different fault states, and obtaining a fault state identification result. And verifying the rationality of the identification method according to a fault state identification result. The multi-scale permutation entropy fault diagnosis method has the advantages that the defect that a conventional multi-scale permutation entropy fault diagnosis algorithm is insufficient in fault feature extraction capacity is overcome, the calculation time is short, and the fault diagnosis efficiency is improved.

Description

technical field [0001] The invention belongs to the field of signal recognition, in particular to a vibration signal recognition method and system based on improved multi-scale permutation entropy. Background technique [0002] At present, with the development of science and technology, the mechanical system is showing a trend of intelligence and integration. The good operation of the mechanical system plays an important role in ensuring the quality and efficiency of industrial development. As a key component in a mechanical system, rolling bearings determine whether their operating status is good or not determines the state performance of the entire mechanical system. Therefore, it is necessary to identify the working status of rolling bearings. Among the existing fault diagnosis methods for rolling bearings based on signal analysis, the diagnosis method based on vibration signals has the characteristics of high frequency of use, simplicity and effectiveness. However, limi...

AI Technical Summary

Problems solved by technology

[0024] (2) According to the characteristics of the conventional multi-scaling method, when the scale factor is τ, the time series x(N) is coarse-grained into a coarse-grained sequence with a length of [N / τ]. For example, when the scale factor τ = 2, the original time series with a length of 2048 becomes a coarse-grained series with a length of 1024, so in the conventional multi-scale analysis process, the length of the time series is shortened with the increase of the scale factor, and there are too short time series The following problems: one is that a lot of information is lost, and the other is that when calculating the permutation entropy, with the shortening of the time series, the parameters need to be adjusted accordingly, which affects the adaptability in the fault diagnosis process

[0025] (3) The fault diagnosis algorithm based on conventional multi-scale permutation entropy has the defect of insufficient ability in fault feature extraction. It is necessary to extract permutation entropy values ​​at multiple scales as feature vectors to effectively summarize feature information. Too much feature information One is that it causes the redundancy of the eigenvectors, and the other is that it increases the calculation time and affects the efficiency of fault diagnosis.

Images