Rotary machine fault diagnosis method based on MF-JADE

Abstract

The invention relates to the technical field of rotary machine fault diagnosis, and concretely relates to a rotary machine fault diagnosis method based on MF-JADE. The method comprises the following steps: 1, collecting rotor vibration signals of a rotary machine through a plurality of sensors; 2, calculating the window width of a median filter according to the rotor vibration signal sampling frequency, and filtering out pulse noise and part of white noise in the signal by adopting median filtering; 3, separating the signal after median filtering by adopting a JADE algorithm to separate out each single fault characteristic signal; and 4, carrying out corresponding diagnosis on each separated single fault characteristic signal. According to the method, the median filtering technology is used for reducing noise of the coupling fault vibration signal of the rotary machine rotor system, so pulse noise and white noise can be effectively filtered out, noise is reduced, and the signal-to-noise ratio is increased; and then blind source separation based on JADE is used for obtaining the optimal effect, and effective extraction of a fault feature signal is achieved, so the method is an effective diagnosis method for the coupling fault of the rotary machine rotor system.

Description

technical field [0001] The invention relates to the technical field of fault diagnosis of rotating machinery, in particular to a method for diagnosing faults of rotating machinery based on MF-JADE. Background technique [0002] In the practice of mechanical equipment (system) condition monitoring and fault diagnosis, the object to be monitored and diagnosed is often not a single mechanical equipment, but multiple mechanical equipment operating at the same time. The signal measured by the vibration sensor installed on the mechanical equipment includes the vibration of other machinery on the one hand, and the vibration generated by its own parts on the other hand. The vibration between different machines and the vibration generated by different parts interfere with each other . [0003] At present, the vibration signals detected by mechanical equipment (system) condition monitoring and fault diagnosis are often the result of multiple vibration source signals and interference ...

AI Technical Summary

Problems solved by technology

However, the diagnostic knowledge currently used is often based on the theoretical analysis conclusions of a certain component, which seriously affects the accuracy of fault diagnosis, and even causes false or missed reports.

Images