Application of a two-dimensional lifting wavelet transform to rotating mechanical vibration data compression

Abstract

This article presents a novel method for rotating mechanical vibration data compression based on a two-dimensional (2D) lifting wavelet transform. This method consists of three stages. First, the periodical vibration data are converted from one-dimensional (1D) to 2D in order to reduce the dependency both within a single cycle and across cycles more efficiently. Then, the 2D matrix is decomposed based on a 2D wavelet transform via biorthogonal filters constructed by the lifting scheme. Second, with the purpose of data compression, the significant wavelet coefficients are selected to retain while the small coefficients are discarded owing for their lesser contribution to the reconstruction signal. Finally, the reserved coefficients are utilized to reconstruct the 2D matrix through inverse wavelet transform and the reconstructed vibration data can be obtained by transforming the matrix from 2D to 1D. To evaluate the performance of the proposed compression method, the real-life rotating mechanical vibration data were used to test this method. The results show that the proposed method can achieve better performance compared to the method based on a 1D wavelet transform.