Ensemble empirical mode decomposition current diagnosis method for motor broken bar faults

Abstract

The invention discloses an ensemble empirical mode decomposition current diagnosis method for motor broken bar faults. According to the characteristics of the motor rotor broken bar faults, the advantages of EEMD decomposition and Hilbert demodulation are combined to adopt the EEMD analysis method with adaptivity and a high signal-to-noise ratio to perform self-adaption decomposition on a motor broken bar current signal in the no-load and load conditions of a motor. The whole frequency band is finely divided, and then an IMF component with a large correlation coefficient with an original signal is selected to reconstruct the signal. Hilbert envelope demodulation is performed on the reconstructed signal to extract fault characteristics related to the faults from the demodulation result of the current signal; and the fault characteristic information is highlighted and separated, so that the characteristics of the motor broken bar faults are more obvious in the current signal. A basis isprovided for current detection and diagnosis of the motor broken bar faults.

Description

technical field [0001] The invention belongs to the technical field of detection, and in particular relates to a collective empirical mode decomposition current diagnosis method for a broken bar fault of an asynchronous induction motor rotor used for state monitoring and fault diagnosis. Background technique [0002] The signal collected by the current sensor is a time-domain waveform. The time-domain waveform is more intuitive and contains a large amount of information, but the relationship with the fault is not obvious. Generally, it cannot be directly used as a basis for judgment, but it can be used as an auxiliary feature. Diagnosis results are more precise. [0003] Spectrum analysis is a commonly used method in signal analysis. Fourier transform is the theoretical basis of spectrum analysis. It is very effective for the analysis of stationary signals, but one disadvantage is that it lacks spatial locality and is not very effective for abrupt signals, such as Fourier tr...

AI Technical Summary

Problems solved by technology

[0003] Spectrum analysis is a commonly used method in signal analysis. Fourier transform is the theoretical basis of spectrum analysis. It is very effective for the analysis of stationary signals, but one disadvantage is that it lacks spatial locality and is not very effective for abrupt signals, such as Fourier transform. Leaf transform is very effective for uniform heavy load current signal analysis, not very effective for no load or variable load

[0004] In 2006, researchers Gaetan Dither, Eric Ternisien and others from t

Images