Rolling bearing intelligent fault diagnosis method based on multi-classification fuzzy correlation vector machine

Abstract

The invention discloses a rolling bearing intelligent fault diagnosis method based on a multi-classification fuzzy correlation vector machine. The acceleration sensor is used for collecting vibrationsignals of the rolling bearing; wavelet packet energy feature vectors of all operation states are obtained under different noise intensities; a new class center-based membership degree calculation method is introduced; based on this, a multi-classification fuzzy correlation vector machine is constructed to realize intelligent fault diagnosis of the rolling bearing; the fault feature vector sampleset is trained and tested by using the multi-classification fuzzy correlation vector machine, and the test result is compared with the actual fault type, so that the effectiveness of the diagnosis method is verified, and the intelligent fault diagnosis of the rolling bearing is realized. The diagnosis method overcomes the defect that a traditional intelligent fault diagnosis method is not high indiagnosis accuracy in a strong noise environment, is high in fault diagnosis efficiency and good in anti-noise performance, is suitable for rolling bearing fault diagnosis in a complex noise environment, and has good engineering value and application prospect.

Description

technical field [0001] The invention belongs to the technical field of intelligent fault diagnosis of rolling bearings, and in particular relates to an intelligent fault diagnosis method of rolling bearings based on a multi-classification fuzzy correlation vector machine. Background technique [0002] The operation status of rolling bearings in large rotating machinery is directly related to the safe and economical operation of the equipment. Once the rolling bearings fail, it will cause a chain reaction, ranging from failure of the transmission system to serious accidents involving machine crashes and fatalities. Therefore, it is of great significance to monitor the running status and fault diagnosis of rolling bearings. [0003] Because the vibration signal of rolling bearing has the advantages of convenient data acquisition and wide applicability, the existing intelligent fault diagnosis method of rolling bearing mainly collects the vibration signal for fault feature extr...

AI Technical Summary

Problems solved by technology

[0031] The present invention uses a one-to-one method to realize a multi-category fuzzy correlation vector machine, and provides a new calculation method for sample membership degree. The multi-classification fuzzy correlation vector machine realizes the intelligent fault diagnosis of rolling bearings. Compared with the existing intelligent fault identification technology of rolling bearings, it overcomes the poor noise resistance and poor diagnostic effect caused by the sensitivity of existing intelligent fault diagnosis methods of rolling bearings to noise points or abnormal points. The shortcoming of the ideal, by introducing a new membership degree calculation method based on the class center, while enhancing the role of the correlation vector, it reduces the adverse impact of the noise point on the classification, and at the same time extracts the wavelet packet energy features of the vibration signal of the rolling bearing, which improves the failure rate. Diagnosis accuracy, efficient identification of rolling bearing fault types

Images