Fault Diagnosis Method of Rolling Bearing Based on MCKD Algorithm and Support Vector Machine

Abstract

The invention discloses a rolling bearing fault diagnosis method based on an MCKD algorithm and a support vector machine. Step 1) Utilizes the characteristics of the maximum correlation kurtosis deconvolution algorithm, takes the maximization of the signal correlation kurtosis as the optimization goal, and can complete the vibration signal through iteration The deconvolution operation highlights the continuous pulse covered by strong noise in the signal, and extracts the characteristic signal representing the vibration characteristics of the rolling bearing in the signal with a low signal-to-noise ratio; step 2) utilizes the classifier constructed by the support vector machine, Complete the training and learning of characteristic signals, and classify them according to categories in the space, each category represents a motion state of rolling bearings, so as to realize the fault diagnosis of rolling bearings according to categories. The support vector machine classification algorithm parameter setting of the invention is simple, the calculation complexity is low, the diagnosis precision is high, the diagnosis result is more direct, and has wide application value in the fields of production and manufacturing.

Description

technical field [0001] The invention relates to the fields of signal processing, pattern recognition, etc., and can analyze, process and identify the vibration signal of the rolling bearing, and finally determine the running state of the bearing according to the vibration signal. Background technique [0002] Rolling bearings are known as the joints of rotating mechanical equipment, and their health directly affects the state of the entire machine. When the bearing is defective due to wear, overload, etc., the equipment will vibrate abnormally and generate noise. In severe cases, the mechanical device will be damaged and economic losses will be caused. Therefore, the fault detection and diagnosis of rolling bearings has important research significance. [0003] Whether the feature extraction of nonlinear vibration signal is accurate and effective in rolling bearing fault diagnosis determines the reliability of diagnosis. However, there are many vibration equipment in the in...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is: it is difficult for the general algorithm to extract the vibration signal with a low signal-to-noise ratio, and the algorithm cannot directly distinguish the fault type and other problems

Images