A novel, machine learning-based feature extraction method for detecting and localizing bearing component defects
Vibration analysis for conditional preventive maintenance is an essential tool for the industry. The vibration signals sensored, collected and analyzed can provide information about the state of an induction motor. Appropriate processing of these vibratory signals leads to define a normal or abnorma...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Polish Academy of Sciences
2022-06-01
|
| Series: | Metrology and Measurement Systems |
| Subjects: | |
| Online Access: | https://journals.pan.pl/Content/123578/PDF/a07.pdf |
| _version_ | 1811221047851089920 |
|---|---|
| author | Bilal Djamal Eddine Cherif Sara Seninete Mabrouk Defdaf |
| author_facet | Bilal Djamal Eddine Cherif Sara Seninete Mabrouk Defdaf |
| author_sort | Bilal Djamal Eddine Cherif |
| collection | DOAJ |
| description | Vibration analysis for conditional preventive maintenance is an essential tool for the industry. The vibration signals sensored, collected and analyzed can provide information about the state of an induction motor. Appropriate processing of these vibratory signals leads to define a normal or abnormal state of the whole rotating machinery, or in particular, one of its components. The main objective of this paper is to propose a method for automatic monitoring of bearing components condition of an induction motor. The proposed method is based on two approaches with one based on signal processing using the Hilbert spectral envelope and the other approach uses machine learning based on random forests. The Hilbert spectral envelope allows the extraction of frequency characteristics that are considered as new features entering the classifier. The frequencies chosen as features are determined from a proportional variation of their amplitudes with the variation of the load torque and the fault diameter. Furthermore, a random forest-based classifier can validate the effectiveness of extracted frequency characteristics as novel features to deal with bearing fault detection while automatically locating the faulty component with a classification rate of 99.94%. The results obtained with the proposed method have been validated experimentally using a test rig. |
| first_indexed | 2024-04-12T07:52:01Z |
| format | Article |
| id | doaj.art-1902d10027574f53ba43e887206bc64f |
| institution | Directory Open Access Journal |
| issn | 2300-1941 |
| language | English |
| last_indexed | 2024-04-12T07:52:01Z |
| publishDate | 2022-06-01 |
| publisher | Polish Academy of Sciences |
| record_format | Article |
| series | Metrology and Measurement Systems |
| spelling | doaj.art-1902d10027574f53ba43e887206bc64f2022-12-22T03:41:34ZengPolish Academy of SciencesMetrology and Measurement Systems2300-19412022-06-01vol. 29No 2333346https://doi.org/10.24425/mms.2022.140038A novel, machine learning-based feature extraction method for detecting and localizing bearing component defectsBilal Djamal Eddine Cherif0Sara Seninete1Mabrouk Defdaf2Department of Electrical Engineering, Faculty of Technology, University of M’sila, M’sila 28000, AlgeriaDepartment of Electrical Engineering, Faculty of Technology, University of Mostaganem, Mostaganem 27000, AlgeriaDepartment of Electrical Engineering, Faculty of Technology, University of M’sila, M’sila 28000, AlgeriaVibration analysis for conditional preventive maintenance is an essential tool for the industry. The vibration signals sensored, collected and analyzed can provide information about the state of an induction motor. Appropriate processing of these vibratory signals leads to define a normal or abnormal state of the whole rotating machinery, or in particular, one of its components. The main objective of this paper is to propose a method for automatic monitoring of bearing components condition of an induction motor. The proposed method is based on two approaches with one based on signal processing using the Hilbert spectral envelope and the other approach uses machine learning based on random forests. The Hilbert spectral envelope allows the extraction of frequency characteristics that are considered as new features entering the classifier. The frequencies chosen as features are determined from a proportional variation of their amplitudes with the variation of the load torque and the fault diameter. Furthermore, a random forest-based classifier can validate the effectiveness of extracted frequency characteristics as novel features to deal with bearing fault detection while automatically locating the faulty component with a classification rate of 99.94%. The results obtained with the proposed method have been validated experimentally using a test rig.https://journals.pan.pl/Content/123578/PDF/a07.pdfinduction motorbearingfaultouter raceinner raceenvrf |
| spellingShingle | Bilal Djamal Eddine Cherif Sara Seninete Mabrouk Defdaf A novel, machine learning-based feature extraction method for detecting and localizing bearing component defects Metrology and Measurement Systems induction motor bearing fault outer race inner race env rf |
| title | A novel, machine learning-based feature extraction method for detecting and localizing bearing component defects |
| title_full | A novel, machine learning-based feature extraction method for detecting and localizing bearing component defects |
| title_fullStr | A novel, machine learning-based feature extraction method for detecting and localizing bearing component defects |
| title_full_unstemmed | A novel, machine learning-based feature extraction method for detecting and localizing bearing component defects |
| title_short | A novel, machine learning-based feature extraction method for detecting and localizing bearing component defects |
| title_sort | novel machine learning based feature extraction method for detecting and localizing bearing component defects |
| topic | induction motor bearing fault outer race inner race env rf |
| url | https://journals.pan.pl/Content/123578/PDF/a07.pdf |
| work_keys_str_mv | AT bilaldjamaleddinecherif anovelmachinelearningbasedfeatureextractionmethodfordetectingandlocalizingbearingcomponentdefects AT saraseninete anovelmachinelearningbasedfeatureextractionmethodfordetectingandlocalizingbearingcomponentdefects AT mabroukdefdaf anovelmachinelearningbasedfeatureextractionmethodfordetectingandlocalizingbearingcomponentdefects AT bilaldjamaleddinecherif novelmachinelearningbasedfeatureextractionmethodfordetectingandlocalizingbearingcomponentdefects AT saraseninete novelmachinelearningbasedfeatureextractionmethodfordetectingandlocalizingbearingcomponentdefects AT mabroukdefdaf novelmachinelearningbasedfeatureextractionmethodfordetectingandlocalizingbearingcomponentdefects |