Separation of vibration components based on sparse nonnegative tensor factorization

Haobin  Wen; Lin  Liang; Ben  Niu; Lei  Shan; Maolin  Li; Guang  Li

Haobin Wen School of Mechanical Engineering, Xi’an Jiaotong University, Xi’an
Lin Liang Key Laboratory of Education Ministry for Modern Design and Rotor-Bearing System, Xi’an Jiaotong University, Xi’an
Ben Niu School of Mechanical Engineering, Xi’an Jiaotong University, Xi’an
Lei Shan School of Mechanical Engineering, Xi’an Jiaotong University, Xi’an
Maolin Li Workshop, Xi’an Jiaotong University, Xi’an
Guang Li School of Mechanical Engineering, Xi’an Jiaotong University, Xi’an

Keywords: Sparse nonnegative tensor factorization, Source separation, Short Time Fourier Transform, Gear fault

Abstract

The vibration signals collected for machinery fault diagnosis tend to contain multiple excitation sources. Independent Component Analysis (ICA), the common approach of Blind Source Separation (BSS) to separate mixed source signals, is challenged when the data distributions are not known in advance. To circumvent such preconditions, the source separation method based on sparse nonnegative tensor factorization (SNTF) is proposed. First, the multichannel vibration signals are constructed into a third-order time-frequency tensor by means of short-time Fourier transform (STFT). Then, the spectrum tensor is decomposed into three factors by performing nonnegative tensor factorization with l1 sparsity constraint, and thus the source signals can be reconstructed separately utilizing different decomposed subspace. Results of the practical experiment on a two-stage gearbox indicate the performance improvement crediting to the sparseness penalty, and therefore verify the effectiveness of the separation method.