Feature Extraction of Vibration Signal Based on an Improved Local Wave Analysis
Feng Li Wang, Hui Lin Xing, Shu Lin Duan · Applied Mechanics and Materials · 2014
In order to reduce the random noise influence on local wave decomposition (LWD), the original data is adaptively denoised by lifting wavelet transform to strain mode aliasing in LWD, avoid the pseudo mode functions in LWD and improve the quality of LWD. Local wave marginal spectrum is employed to analyze vibration signal for extracting the characteristics information. Intrinsic mode functions obtained by LWD, the instantaneous frequency and amplitude can be calculated by the Hilbert transform. The improved local wave analysis can offer a measure of the energy distribution of vibration signal from its instantaneous frequency. Analysis and experimental result demonstrate the validity of the proposed method.