Device vibration signal feature extraction method based on EEMD-CWD

Abstract

The invention discloses a device vibration signal feature extraction method based on EEMD-CWD, comprising the steps of collecting a device vibration signal; performing EEMD of the vibration signal to obtain a group of IMF components; based on the kurtosis criterion, optimizing the IMF components; and extracting fault feature information of the device through a CWD analysis. The device vibration signal feature extraction method is advantageous in that, an EEMD method is introduced, and after white noise is added to an original signal, via using the statistics characteristic of uniform distribution of white noise frequency, an intermittence phenomenon of the original signal can be eliminated, and therefore, modal aliasing can be inhibited; and also, the vibration signal can be decomposed into several IMF components with relatively single frequency, and through the CWD technical analysis of a particular IMF, the effect of reducing frequency aliasing and interference can be achieved.

Description

technical field [0001] The invention belongs to the field of equipment maintenance, in particular to a method for extracting equipment vibration signal features based on EEMD-CWD. Background technique [0002] The feature extraction of equipment vibration information is the process of extracting useful information from the original signal of the equipment, and the purpose is to extract the feature value that can reflect the health status information of the equipment. Therefore, the extraction of equipment vibration signal features is of great significance for monitoring whether the equipment is healthy and working. There are many methods for feature extraction of equipment vibration information, but there are more or less problems in all of them. [0003] Prior art one: [0004] In recent years, the Hilbert-Huang Transform (HHT) has been widely used in the field of signal feature extraction. The main content of HHT consists of two parts. The first part is Empirical Mode D...

AI Technical Summary

Problems solved by technology

[0012] The object of the present invention is: the present invention provides a kind of equipment vibration signal characteristic extraction method based on EEMD-CWD, by introducing integrated empirical mode decomposition method (Ensemble empirical mode decomposition, EEMD) and combining CWD analysis method, solve above-mentioned prior art Existing problems such as modal aliasing and suppression effect are not obvious

Images