Nonlinear sparse mode decomposition and its application in planetary gearbox fault diagnosis

• Published in: Mechanism and machine theory Vol. 155; p. 104082
• Main Authors: Pan, Haiyang, Zheng, Jinde, Yang, Yu, Cheng, Junsheng
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.01.2021
• Subjects: Fault diagnosis; Nonlinear sparse mode decomposition; Planetary gearbox; Singular local linear operator; Fault diagnosis; Singular local linear operator; Planetary gearbox; Nonlinear sparse mode decomposition
• ISSN: 0094-114X; 1873-3999
• DOI: 10.1016/j.mechmachtheory.2020.104082

•A nonlinear sparse mode decomposition (NSMD) is proposed.•The proposed NSMD method can avoid modal aliasing and improve the orthogonality of decomposition.•NSMD method has good robustness and adaptability. Traditional time-frequency analysis methods, including empirical mode decomposition (EMD), local characteristic-scale decomposition (LCD) and variable mode decomposition (VMD), have some limitations in nonlinear signal analysis. When the signal has strong noise, traditional time-frequency analysis methods will force the signal to be decomposed into several inaccurate components, and the achieved components usually suffer from the end effect problem. Considering the above pressing challenge, a new signal decomposition algorithm, nonlinear sparse mode decomposition (NSMD), is proposed in this protocol. The core of NSMD is that the local narrowband signal disappears under the action of the singular local linear operator, so the singular local linear operator can be applied to extract the local narrowband component of the detected signal. Meanwhile, the obtained local narrowband signal can be superposed as the basic signal to close to the original signal, realizing the adaptive decomposition of the signal with good robustness and adaptability. The analysis results of simulation signals and planetary gearbox fault signals indicate that the proposed NSMD method is effective for raw vibration signals.