Adaptive power spectrum Fourier decomposition method with application in fault diagnosis for rolling bearing. (October 2021)

• Record Type: Journal Article
• Authors: Zheng, Jinde; Huang, Siqi; Pan, Haiyang; Tong, Jinyu; Wang, Chengjun; Liu, Qingyun
• Is Part Of: Measurement. Volume 183(2021)
• Publication Date: 2021-10
• Subjects: Fourier decomposition method -- Adaptive power spectrum Fourier decomposition method -- Empirical wavelet transform -- Rolling bearing -- Fault diagnosis; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2021.109837
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: A newly adaptive power spectrum Fourier decomposition method (APSFDM) is proposed. The decomposition of APSFDM is adaptive and complete. Simulation signal analysis verified the superiority of APSFDM in signal fidelity. APSFDM gets a better diagnostic effect than other compared methods. Abstract: Fourier decomposition method (FDM) is a recently proposed method for non-stationary signal decomposition. In FDM, the conditions for determining mono-components are easy to meet, which will lead to over-decomposition of signals. In this paper, a new adaptive power spectrum Fourier decomposition method (APSFDM) is proposed based on FDM. First, the intervals where the components located in the power spectrum of raw signal are adaptively searched and then the signal is decomposed into several mono-components with physically meaningful instantaneous frequencies by a reconstruction way. The APSFDM method is compared with existing empirical wavelet transform (EWT), empirical mode decomposition (EMD), variational mode decomposition (VMD) and FDM methods through simulation signal analysis to verify its superiority in signal fidelity. Finally, APSFDM method is employed to the fault diagnosis of rolling bearing with comparison it with the above mentioned existing method. The analysis results of the measured bearing data indicate that the mono-components obtained by APSFDM contain more accurate fault feature information that can be used for an effective failure diagnosis of bearing. The effectiveness of APSFDM method is further verified by comparing its decomposition and diagnostic results with existing methods.