Open AccessCharacteristic signal based on the combination of empirical mode decomposition method and time series AR model Extraction method
Author(s) -
Binbin Li,
Xiangdong Li,
Pu Xu-Yang,
Kang Zhong-Tao,
Zongyang Li,
Zhang Hui-Zhuo,
Chenxi Chen,
Hu Y
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1986/1/012126
Subject(s) - hilbert–huang transform , wavelet , autoregressive model , signal (programming language) , algorithm , series (stratigraphy) , wavelet transform , wavelet packet decomposition , matrix (chemical analysis) , mathematics , spectral density , computer science , artificial intelligence , statistics , white noise , paleontology , materials science , composite material , biology , programming language
In the process of signal decomposition by wavelet theory, the wavelet basis function is artificially selected based on experience, and the method based on empirical mode decomposition is decomposed according to the time scale of the signal itself. The article uses two methods to decompose a certain segmented frequency conversion signal to obtain the intrinsic modal component matrix and the wavelet decomposition coefficient matrix, then calculates the Hilbert time spectrum of the two decomposition matrices. cThe calculations shows that the false information generated by the empirical mode decomposition signal is obviouly more less. Therefore, the empirical mode decomposition method is used to decompose the bearing vibration signal, and the stationary natural mode function obtained is very suitable for establishing an autoregressive AR model to extract the power spectrum of each component for analysis. Finally, the characteristic frequencies of different states of rolling bearings are extracted, which provides support for data-driven fault diagnosis.