A Fault Diagnosis Method for Rolling Bearings Based on Improved Chebyshev Distance

Abstract

This application discloses a rolling bearing fault diagnosis method and system based on the improved Chebyshev distance, select the sample data of the normal working condition and the fault working condition in the rolling bearing, divide each set of sample data of 10 sets of sample data into 10 equal parts; and then carry out EMD decomposition, and select the first 5 IMF components after decomposition; perform SDP transformation on each IMF component, binarize the SDP image, select the local matrix of the image, and then calculate the mean matrix of 10 local matrices; for each group of samples Salt and pepper denoising of the local matrix and the mean matrix; calculate the largest eigenvalue root of the mean matrix; calculate the improved Chebyshev distance between the 10 matrices in each group of samples and their mean matrix; select 10 groups of sample data to improve the maximum Chebyshev distance value and the minimum value to determine which working condition the test data belongs to. The improved Chebyshev distance can more accurately distinguish the normal state of rolling bearings, ball faults, outer ring faults, and inner ring faults.

Description

technical field [0001] The present application relates to the technical field of rolling bearing operation fault diagnosis, and in particular to a rolling bearing fault diagnosis method based on improved Chebyshev distance. Background technique [0002] Rolling bearings are commonly used rotating machinery in industry. Whether they can operate normally is related to the safety of the entire industrial equipment. Therefore, it is particularly important to detect and diagnose rolling bearings. [0003] The methods of time-frequency analysis are commonly used at home and abroad for fault diagnosis of rolling bearings, such as wavelet transform, Fourier decomposition, etc., but these methods have certain limitations. For example, wavelet transform needs to select wavelet bases. Different wavelet bases lead to different results of wavelet decomposition. However, there is no better method for selecting wavelet bases. In the existing fault diagnosis technology, there are still pro...

AI Technical Summary

Problems solved by technology

In the existing fault diagnosis technology, there are still problems such as the extracted feature quantity is not accurate enough, and the recognition rate is low.

[0004] At present, the difficulty in fault diagnosis of rolling bearings lies in feature extraction, because the collected vibration signals of rolling bearings generally have strong non-Gaussian and nonlinear characteristics, and it is difficult to distinguish the state of bearings with traditional time domain features and frequency domain features.

Images