A demodulating approach based on local mean decomposition and its applications in mechanical fault diagnosis

Abstract

Since machinery fault vibration signals are usually multicomponent modulation signals, how to decompose complex signals into a set of mono-components whose instantaneous frequency (IF) has physical sense has become a key issue. Local mean decomposition (LMD) is a new kind of time–frequency analysis approach which can decompose a signal adaptively into a set of product function (PF) components. In this paper, a modulation feature extraction method-based LMD is proposed. The envelope of a PF is the instantaneous amplitude (IA) and the derivative of the unwrapped phase of a purely flat frequency demodulated (FM) signal is the IF. The computed IF and IA are displayed together in the form of time–frequency representation (TFR). Modulation features can be extracted from the spectrum analysis of the IA and IF. In order to make the IF have physical meaning, the phase-unwrapping algorithm and IF processing method of extrema are presented in detail along with a simulation FM signal example. Besides, the dependence of the LMD method on the signal-to-noise ratio (SNR) is also investigated by analyzing synthetic signals which are added with Gaussian noise. As a result, the recommended critical SNRs for PF decomposition and IF extraction are given according to the practical application. Successful fault diagnosis on a rolling bearing and gear of locomotive bogies shows that LMD has better identification capacity for modulation signal processing and is very suitable for failure detection in rotating machinery.

Wind Turbine Diagnosis under Variable Speed Conditions Using a Single Sensor Based on the Synchrosqueezing Transform Method

“…The LMD can decompose such signals into several layers. The noise can influence the effective layers decomposition order [36]. In order to illustrate the signal with strong noise and investigate the robustness of LMD method to noise, a multi-component AM-FM signal x ( t ) is considered as follows: $xfalse(tfalse)=\cos false(200\mathrm{sans-serif\pi }t+1.5\sin false(18\mathrm{sans-serif\pi }tfalse)false)\cdot false(1+0.5\cos false(14\mathrm{sans-serif\pi }tfalse)false)$…”

A new rotating machinery fault diagnosis method based on improved local mean decomposition

“…For a given signal z(t) = a(t)e jφ(t) , we define the instantaneous bandwidth of the signal z(t) at time t is B t = á(t)/a(t) (15) The instantaneous bandwidth is used to estimate how the P (t, w) concentrates in the center IF at time t. Moreover, when the B t is very small, z(t) is considered as local narrowband signal and has perfect physical sense. Suppose that B is the bandwidth of z(t), we can obtain Eq.…”

“…Cheng et al [14] applied LMD method to fulfill the gear and roller bearing fault diagnosis. Also LMD method is employed to complete the fault diagnosis on a rolling bearing and gear of locomotive bogies by Chen et al [15].…”

A Rolling Bearing Fault Diagnosis Method Based on Variational Mode Decomposition and an Improved Kernel Extreme Learning Machine

“…Inspired by EMD, Smith proposed another adaptive signal decomposition method, local mean decomposition (LMD), in 2005 [7]; this method also attracted significant attention from researchers, and numerous LMD-based diagnosis methods were proposed in succession. Chen et al [8] employed LMD for rolling bearing and gear fault diagnosis. Liu et al [9] obtained a wind power generator's vibration signal instantaneous frequency via LMD to monitor the wind power generator's state.…”