Bearing fault diagnosis method using probabilistic principal component analysis to enhance cyclic bispectrum

Abstract

The invention belongs to the field of mechanical equipment fault diagnosis. The invention discloses a bearing fault diagnosis method using a probabilistic principal component to enhance a cyclic bispectrum. The method includes the steps of firstly, establishing a probabilistic principal component model capable of reflecting a potential relation between an original signal and a principal componentvector; secondly, de-noising the original signal by using the probabilistic principal component model to completely preserve useful components of the signal, greatly increase the signal-to-noise ratioand solve the problem that a fault signal is susceptible to noise interference; and thirdly, subjecting the de-noised signal to cyclic bispectrum analysis, and diagnosing a bearing fault from a contour map of the single cyclic frequency bispectrum. For one thing, by using the probabilistic principal component analysis for de-noising, the method of the invention can effectively increase the signal-to-noise ratio and suppress the noise to highlight a phenomenon of fault frequency modulation near the bearing resonance frequency, thereby increasing the definition of a hexagonal structure formed by the cyclic bispectrum; and for another, can visually detect the type of bearing fault from the contour map of the single cyclic frequency bispectrum.

Description

technical field [0001] The invention belongs to the technical field of mechanical equipment fault diagnosis, in particular to a bearing fault diagnosis method using probability principal component analysis to enhance cycle bispectrum. Background technique [0002] Rotating machinery is generally composed of supporting parts (bearings), rotating parts (shafting) and transmission parts (gears). It is an indispensable and important power device for modern equipment. Its structure is complex and diverse, and it is widely used in modern In the field of mechanical processing, nuclear power, wind energy, hydropower energy, large unit equipment and aviation equipment. According to statistics, among the failures of rotating machinery, bearings are vulnerable parts, and their failures account for 19% of rotating machinery failures. Therefore, it is very valuable and practical to monitor the working state of the bearing in real time so as to detect abnormal faults early. [0003] How...

AI Technical Summary

Problems solved by technology

[0006] In order to achieve the above object, the technical solution adopted by the present invention is: a bearing fault diagnosis method using probability principal component analysis to enhance cyclic bispectrum, which is characterized in that it includes the following steps: ① Establishing a potential relationship that can reflect the original signal and the principal component vector Probabilistic principal component model; ② Use the probability principal component model to denoise the original signal to obtain a high signal-to-noise ratio signal, completely retain the useful components of the signal, and greatly improve the signal-to-noise ratio to solve the problem that the faulty signal is susceptible to noise interference Problem; ③The cyclic bispectrum analysis is performed on the denoising signal, and the contour map of the single cyclic frequency bispectrum is obtained by calculating the third-order cumulant, sine extraction operation, and two-dimensional FFT transformation. The regular hexagonal vertex coordinates in the high-line diagram are compared with the fault characteristic frequency, and the bearing fault is finally diagnosed;

Images