| Gear fault identification and classification of singular value decomposition based on Hilbert-Huang transform
Zhongyuan Su*, Yaoming Zhang, Minping Jia, Feiyun Xu, Jianzhong Hu
The Journal of Mechanical Science and Technology, vol. 25, no. 2, pp.267-272, 2011
Abstract : An improved singular value decomposition method of gear fault identification based on Hilbert-Huang transform was proposed to
overcome the problem of reconstructing a feature matrix of singular value decomposition. The method includes three steps. First, the
instantaneous frequency and amplitude matrices were acquired by Hilbert-Huang transform from faulted gear signals. Second, after the
matrices were decomposed by singular value decomposition, the defined distances of singular value vectors and the optimal threshold of
the distance for classification were calculated. Third, the fault characteristics of a gearbox were identified and classified by the threshold
of the distances. The result demonstrates that the proposed method effectively identifies the gear fault and can realize an automatic gear
fault diagnosis.
Keyword : Hilbert-Huang transform; Time-frequency analysis; Singular value decomposition; Gear fault diagnosis |
