Ultrahigh frequency noise assisted empirical mode decomposition (EMD) method

Abstract

The invention relates to the technical field of mechanical fault diagnosis, and particularly relates to an ultrahigh frequency noise assisted empirical mode decomposition (EMD) method. Ultrahigh frequency noise is added once to perform EMD treatment, the number of screening times is increased when intrinsic mode function (IMF) is extracted by using the characteristic that the ultrahigh frequency noise has many extreme points, and thus a plurality of transient components close to separation frequency spectrum are separated. As the UNEMD is only subjected to EMD once, the calculation amount is relatively small; and the ultrahigh frequency noise is added, so that the acquired IMFs do not include residual noise.

Description

technical field [0001] The invention relates to the technical field of mechanical fault diagnosis, in particular to an ultra-high frequency noise-assisted empirical mode decomposition method. Background technique [0002] Rolling bearings are common parts of rotating machinery and equipment, and they are also vulnerable parts. If the rolling bearing failure is allowed to develop, it will damage other connected components, or even the entire mechanical system, which may eventually cause huge economic losses and adverse social impacts on production or life. Therefore, it is of great significance to the condition monitoring and fault diagnosis of rolling bearings. After the rolling bearing fails, the roller will periodically collide with the faulty part when it rotates. This periodic signal will modulate the resonance frequency of the mechanical system, and it will appear as a periodic transient component in the vibration time domain signal. , appear as equally spaced resonan...

AI Technical Summary

Problems solved by technology

[0009] The present invention aims to solve the above defects, and provides a UHF noise-assisted empirical mode decomposition method, which aims at the problems of large amount of calculation, residual noise and poor separation ability of multiple transient components in the EEMD method. EMD processing is performed after adding ultra-high frequency noise at one time, and the number of screening times during IMF extraction is increased by using the characteristics of ultra-high frequency noise extreme points, so as to separate multiple transient components adjacent to the frequency spectrum

Images