Rotary machine composite fault diagnosis method based on group decomposition

Abstract

The invention discloses a rotary machine composite fault diagnosis method based on group decomposition, comprising a step S1 of measuring the gear box of a rotary machine by using an acceleration sensor to obtain a vibration acceleration original signal x(n); a step S2 of subjecting the vibration acceleration original signal x(n) to group decomposition to obtain an oscillation component OCm(n); astep S3 of performing Hilbert envelope demodulation on the oscillation component OCm(n) to obtain an envelope spectrum Xocm(f) of the oscillation component Xocm(f); a step S4 of identifying whether apreset fault feature frequency and its multiplied frequency are included in the envelope spectrum Xocm(f). The method decomposes the vibration acceleration signal of the rotary machine by using the group decomposition method, adaptively decomposes a non-stationary multi-component vibration signal into a plurality of single-mode oscillation components with instantaneous frequencies having physicalsignificance, performs envelope demodulation on the oscillation components to obtain the component envelope spectrum, analyzes the envelope spectrum, determines the faulty parts and fault types, and accurately diagnose the faults.

Description

technical field [0001] The invention relates to the technical field of mechanical fault diagnosis, in particular to a compound fault diagnosis method for rotating machinery based on group decomposition. Background technique [0002] The failure of mechanical equipment with gears and bearings generally has periodic pulse impact force, resulting in modulation of vibration signals. The frequency spectrum of the modulated signal contains a wealth of fault information. The demodulation analysis method is used to extract the modulation information from the signal, and the degree and type of the fault of the part can be judged by analyzing its intensity and frequency. The Hilbert demodulation method is currently the most commonly used method for vibration signal demodulation analysis. This method can study the amplitude envelope, instantaneous phase and instantaneous frequency of the signal. But the Hilbert change requires the signal to have the characteristic of narrowband and si...

AI Technical Summary

Problems solved by technology

The EMD method decomposes the nonlinear and unstable signal into the sum of several intrinsic mode functions (IMF) with physical meaning, which can be adaptively decomposed according to the local time-varying characteristics of the signal, but there is no suitable mathematical model for EMD , lacks strict mathematical basis, and has defects such as endpoint effects and mode aliasing

At the same time, the EMD component has a large bandwidth and cannot decompose two frequency components with close frequencies

[0003] It can be seen that in the prior art, there is no effective and accurate method to diagnose the composite fault of rotating machinery

Images