Compound Fault Feature Extraction of Rolling Bearing Acoustic Signals Based on AVMD-IMVO-MCKD
The compound fault acoustic signal of a rolling bearing has the characteristics of a varying noise mixture, a low signal-to-noise ratio (SNR), and nonlinearity, which makes it difficult to separate and extract exactly the fault features of compound fault signals. A fault feature extraction approach...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/18/6769 |
| _version_ | 1827656892561752064 |
|---|---|
| author | Shishuai Wu Jun Zhou Tao Liu |
| author_facet | Shishuai Wu Jun Zhou Tao Liu |
| author_sort | Shishuai Wu |
| collection | DOAJ |
| description | The compound fault acoustic signal of a rolling bearing has the characteristics of a varying noise mixture, a low signal-to-noise ratio (SNR), and nonlinearity, which makes it difficult to separate and extract exactly the fault features of compound fault signals. A fault feature extraction approach combining adaptive variational modal decomposition (AVMD) and improved multiverse optimization (IMVO) algorithm parameterized maximum correlated kurtosis deconvolution (MCKD)—named AVMD-IMVO-MCKD—is proposed. In order to adaptively select the parameters of VMD and MCKD, an adaptive optimization method of VMD is proposed, and an improved multiverse optimization (IMVO) algorithm is proposed to determine the parameters of MCKD. Firstly, the acoustic signal of bearing compound faults is decomposed by AVMD to generate several modal components, and the optimal modal component is selected as the reconstruction signal depending on the minimum information entropy of the modal components. Secondly, IMVO is utilized to select the parameters of MCKD, and then MCKD processing is performed on the reconstructed signal. Finally, the compound fault features of the bearing are extracted by the envelope spectrum. Both simulation analysis and acoustic signal experimental data analysis show that the proposed approach can efficiently extract the acoustic signal fault features of bearing compound faults. |
| first_indexed | 2024-03-09T22:35:41Z |
| format | Article |
| id | doaj.art-56353c420e2e4e5ebad1c646fd927499 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T22:35:41Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-56353c420e2e4e5ebad1c646fd9274992023-11-23T18:49:00ZengMDPI AGSensors1424-82202022-09-012218676910.3390/s22186769Compound Fault Feature Extraction of Rolling Bearing Acoustic Signals Based on AVMD-IMVO-MCKDShishuai Wu0Jun Zhou1Tao Liu2Faculty of Mechanical and Electrical Engineering, Kunming University of Science and Technology, Kunming 650500, ChinaFaculty of Mechanical and Electrical Engineering, Kunming University of Science and Technology, Kunming 650500, ChinaFaculty of Mechanical and Electrical Engineering, Kunming University of Science and Technology, Kunming 650500, ChinaThe compound fault acoustic signal of a rolling bearing has the characteristics of a varying noise mixture, a low signal-to-noise ratio (SNR), and nonlinearity, which makes it difficult to separate and extract exactly the fault features of compound fault signals. A fault feature extraction approach combining adaptive variational modal decomposition (AVMD) and improved multiverse optimization (IMVO) algorithm parameterized maximum correlated kurtosis deconvolution (MCKD)—named AVMD-IMVO-MCKD—is proposed. In order to adaptively select the parameters of VMD and MCKD, an adaptive optimization method of VMD is proposed, and an improved multiverse optimization (IMVO) algorithm is proposed to determine the parameters of MCKD. Firstly, the acoustic signal of bearing compound faults is decomposed by AVMD to generate several modal components, and the optimal modal component is selected as the reconstruction signal depending on the minimum information entropy of the modal components. Secondly, IMVO is utilized to select the parameters of MCKD, and then MCKD processing is performed on the reconstructed signal. Finally, the compound fault features of the bearing are extracted by the envelope spectrum. Both simulation analysis and acoustic signal experimental data analysis show that the proposed approach can efficiently extract the acoustic signal fault features of bearing compound faults.https://www.mdpi.com/1424-8220/22/18/6769adaptive variational mode decompositionimproved multiverse optimization algorithmmaximum correlated kurtosis deconvolutionbearing compound faultacoustic diagnosis |
| spellingShingle | Shishuai Wu Jun Zhou Tao Liu Compound Fault Feature Extraction of Rolling Bearing Acoustic Signals Based on AVMD-IMVO-MCKD Sensors adaptive variational mode decomposition improved multiverse optimization algorithm maximum correlated kurtosis deconvolution bearing compound fault acoustic diagnosis |
| title | Compound Fault Feature Extraction of Rolling Bearing Acoustic Signals Based on AVMD-IMVO-MCKD |
| title_full | Compound Fault Feature Extraction of Rolling Bearing Acoustic Signals Based on AVMD-IMVO-MCKD |
| title_fullStr | Compound Fault Feature Extraction of Rolling Bearing Acoustic Signals Based on AVMD-IMVO-MCKD |
| title_full_unstemmed | Compound Fault Feature Extraction of Rolling Bearing Acoustic Signals Based on AVMD-IMVO-MCKD |
| title_short | Compound Fault Feature Extraction of Rolling Bearing Acoustic Signals Based on AVMD-IMVO-MCKD |
| title_sort | compound fault feature extraction of rolling bearing acoustic signals based on avmd imvo mckd |
| topic | adaptive variational mode decomposition improved multiverse optimization algorithm maximum correlated kurtosis deconvolution bearing compound fault acoustic diagnosis |
| url | https://www.mdpi.com/1424-8220/22/18/6769 |
| work_keys_str_mv | AT shishuaiwu compoundfaultfeatureextractionofrollingbearingacousticsignalsbasedonavmdimvomckd AT junzhou compoundfaultfeatureextractionofrollingbearingacousticsignalsbasedonavmdimvomckd AT taoliu compoundfaultfeatureextractionofrollingbearingacousticsignalsbasedonavmdimvomckd |