Local mean decomposition method for fault diagnosis of low-speed rotating machine

Abstract

The invention discloses a local mean decomposition method for fault diagnosis of a low-speed rotating machine. The method is characterized by improving an endpoint effect and a PF screening stop criterion in the local mean decomposition method, and carrying out double-speed processing of a sampled signal so as to effectively avoid a low-frequency-band error during later spectrum analysis; adding an endpoint continuation mechanism of an extreme point in local mean decomposition so that the influence of an endpoint effect can be effectively reduced; by combining a kurtosis criterion and the PF screening stop criterion of pure frequency modulation judgment, improving stability of PF calculation so that a non-convergence phenomenon is not easy to occur; and realizing noise elimination througha small amount of PF function reconstruction. Energy of fault characteristic components can be effectively improved and the energy is more obvious, fault judgment is facilitated, a calculated amount is small, signal noise reduction and fault characteristic analysis are achieved, and the method is suitable for diagnosis of low-speed equipment on a production site.

Description

technical field [0001] The invention relates to a mean value decomposition method for fault diagnosis, specifically a local mean value decomposition method for performing precise vibration spectrum analysis on key components such as bearings and gearboxes of low-speed rotating machines such as mixers and low-speed conveyors to achieve fault analysis. Technical field of mechanical fault detection and diagnosis. Background technique [0002] In the current research on fault diagnosis of rotating machinery, machine learning methods are mainly used for intelligent adaptive diagnosis, and deep learning methods that have been widely used in recent years have stronger fault classification performance. The local mean decomposition method was proposed by Jonathan S.Smith in 2005. It can adaptively decompose the vibration signal for further noise reduction and spectrum analysis. This method has been applied in the fault diagnosis of rotating machinery. However, in the diagnosis of lo...

AI Technical Summary

Problems solved by technology

[0003] In view of the above problems, the present invention provides a local mean decomposition method for fault diagnosis of low-speed rotating machinery. By improving the endpoint effect and PF screening stop criterion in the local mean decomposition method, more accurate spectrum analysis and diagnosis can be realized. So as to solve the problem that affects the accuracy of screening

Images