Fault feature extraction of wind turbine gearbox under variable speed based on improved adaptive variational mode decomposition

Abstract

This paper proposes a study on gearbox fault feature extraction of wind turbine under variable speed condition using improved adaptive variational mode decomposition (VMD). Frequent changes in wind speed and critical noise interference make the vibration signal exhibit non-stationary characteristics. Although computational order tracking can transform non-stationary signals in the time domain into stationary angular signals, and then extract fault features from order spectrum obtained by FFT, the obtained order spectrum is liable to be polluted by noise or to appear order aliasing. To avoid order aliasing and eliminate noise interference, the original non-stationary signal is firstly processed using the improved adaptive VMD method called adaptive differential evolution – VMD (ADE-VMD). ADE-VMD can not only utilise the advantages of traditional VMD but also adaptively select narrow-band intrinsic mode function (NBIMF) to construct the reconstructed signal with less noise and without order aliasing. In the experiment, we compared the ADE-VMD method with other VMD methods such as GA-VMD, PSO-VMD and DE-VMD, and the results showed that ADE-VMD has excellent adaptive processing ability, and its convergence and optimisation speed are more remarkable. ADE-VMD can effectively filter the noise inference and avoid the order aliasing, so it is well suitable for fault feature extraction under variable speed.

A Review on Vibration-Based Fault Diagnosis Techniques for Wind Turbine Gearboxes Operating Under Nonstationary Conditions

“…EMD is an adaptive method, applied to decompose nonlinear and nonstationary signals into different components based on oscillations contained. Similar to EMD, variational mode decomposition (VMD) also decomposes a multicomponent signal into band-limited mode functions non-recursively [130]. These band-limited mono-components have been employed for planetary gearbox fault diagnosis using demodulation analysis [131,132].…”

Research on Wind Power Short-Term Forecasting Method Based on Temporal Convolutional Neural Network and Variational Modal Decomposition

“…Compared with EMD, VMD can achieve modal decomposition adaptively and avoid the problem of modal mixing. In recent years, scholars at home and abroad have conducted a series of studies on the decomposition of time-series signals with VMD to reduce the influence of random noise in time-series signals on the time-series prediction model [ 18 , 19 , 20 , 21 ].…”

“…Scholars have shown that the variation of wind field energy is primarily related to environmental factors, but there is significant uncertainty and randomness in these data. With the development of signal analysis theory, empirical modal decomposition (EMD) and variational modal decomposition (VMD) have become the main methods in the industrial field to solve the degradation of model feature learning ability due to the randomness of data [ 18 , 19 , 20 , 21 ]. Among them, the research results of some scholars showed [ 19 , 21 ] that VMD can better solve the modal mixing and frequency adaption problems compared with EMD.…”

Gearboxes fault detection under operation varying condition based on MODWPT, Ant colony optimization algorithm and Random Forest classifier

“…6 Hun et al 7 introduced a deep belief network (DBN) algorithm for gear fault diagnosis based on wavelet packet energy entropy (WPEE) and multi-scale permutation entropy (MPE). Zheng et al 8 worked on the extraction of gearbox fault feature of wind turbine under variable speed condition using improved adaptive variational mode decomposition (VMD). In the research work published by Gougam et al 9 the decomposition of the initial signal into different modes is done using EWT.…”