Non-Stationary Dynamics Data Analysis With Wavelet-SVD Filtering

Modal identification based on Hilbert-Huang Transform of structural response with SVD preprocessing

“…The technique is based on SVD of the trajectory matrix which is the embedding process of a time series [19]. The trajectory matrix D can be generated from the measured discrete time series x(k) (k = 1, 2, .…”

Gear fault identification and classification of singular value decomposition based on Hilbert-Huang transform

“…When a small perturbation is added, a large variance of its singular value does not occur. However, for some signals, a different signal may produce the same singular value which makes classification difficult [6][7][8]. The proposed technique is based on the SVD for reducing noise from a signal's time series using a timefrequency distribution.…”

Dynamic unbalance detection of Cardan shaft in high-speed train applying double decomposition and double reconstruction method

“…While effective for small quantities, wavelet decomposition of large data volumes often leads to difficulty in judging the data processing result in practice. Therefore, post-processing methods of wavelet decomposition such as wavelet scale energy statistical analysis [6][7][8], wavelet fractal analysis [9][10][11][12], and wavelet singular value decomposition [13][14][15][16] have been fully developed. Because wavelet singular value decomposition has unique advantages in de-noising and eliminating correlations of signals, it has widely been used in fault diagnosis and image processing.…”

“…The complete course includes four steps. Firstly, the vibration acceleration is decomposed into different scale wavelet coefficients, which were traditionally used to construct a matrix to compute its singular value [13][14][15][16]. Secondly, the single scale vibration signal is reconstructed via inverse wavelet transformations of single scale wavelet coefficients.…”