Wavelet-packet-and-support-vector-machine-based on-line detection method for assembling state of main shaft bearing of machine tool

Abstract

The invention relates to a wavelet-packet-and-support-vector-machine-based on-line detection method for an assembling state of a main shaft bearing of a machine tool. The method comprises: step 1, a vibration acceleration sensor is used for collecting vibration signals of a rolling bearing in a normal operation state and a non-normal operation state; step 2, noise elimination processing is carried out on the collected vibration signal and signal wavelet packet reconstruction is carried out on the vibration signals after noise elimination; step 3, m-layer wavelet packet decomposition is carried out on the reconstructed vibration signals, wherein a decomposition signal in each frequency band represents vibration information of the reconstructed signal in a frequency range; step 4, normalization processing is carried out on feature vectors corresponding to all frequency band energy values and a new feature vector is constructed; and step 5, on the basis of the new feature vector, the new feature vector serving as a sample is inputted into a classifier established by using a support vector machine to carry out training, thereby obtaining a trained SVM diagnosis model. Therefore, on-line detection and diagnosis are carried out on an assembling state of a main shaft bearing.

Description

technical field [0001] The invention relates to a method for detecting the assembly quality of a machine tool spindle bearing, in particular to an online detection method for the assembly state of a machine tool spindle bearing based on wavelet packets and support vector machines. Background technique [0002] Since the spindle of a machine tool is a precision system composed of many components assembled, the assembly quality of each component directly affects the service performance of the spindle system, and slight changes in the assembly state of the components in the spindle will have a significant impact on the performance of the spindle . Among them, the bearing is an important part in the spindle system, and its assembly state has a great influence on the performance of the spindle. For example, when the bearing installation is deflected or misplaced, it will cause the spindle to generate noise and abnormal vibration, and the serious one will cause the loss of the pr...

AI Technical Summary

Problems solved by technology

However, these methods are only suitable for dealing with obvious bearing fault signals, and the vibration signal of the main shaft caused by bearing deflection or misalignment does not cause obvious fault features, thus limiting its application.

[0005] In addition, there is no clear standard for the extraction of vibration features of spindle bearings. In order to meet the requirements of fault diagnosis accuracy, multiple fault features usually need to be extracted, which increases the amount of calculation and will limit the accuracy of fault diagnosis with the increase of fault features.

Aiming at the weak signals existing in spindle bearing assembly status recognition, it is difficult to obtain a large number of data samples, recognition knowledge acquisition is difficult, reasoning ability is weak, and feature extraction is difficult. It is the focus of bearing assembly state detection research to quickly determine the assembly state of the spindle bearing by using a low-precision assembly state detection model.

Images