Probability-Based Algorithm for Bearing Diagnosis with Untrained Spall Sizes
Bearing spall detection and predicting its size are great challenges. Model-based simulation is a well-known traditional approach to physically model the influence of the spall on the bearing. Building a physical model is challenging due to the bearing complexity and the expert knowledge required to...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/5/1298 |
| _version_ | 1811303271018528768 |
|---|---|
| author | Ido Tam Meir Kalech Lior Rokach Eyal Madar Jacob Bortman Renata Klein |
| author_facet | Ido Tam Meir Kalech Lior Rokach Eyal Madar Jacob Bortman Renata Klein |
| author_sort | Ido Tam |
| collection | DOAJ |
| description | Bearing spall detection and predicting its size are great challenges. Model-based simulation is a well-known traditional approach to physically model the influence of the spall on the bearing. Building a physical model is challenging due to the bearing complexity and the expert knowledge required to build such a model. Obviously, building a partial physical model for some of the spall sizes is easier. In this paper, we propose a machine-learning algorithm, called Probability-Based Forest, that uses a partial physical model. First, the behavior of some of the spall sizes is physically modeled and a simulator based on this model generates scenarios for these spall sizes in different conditions. Then, the machine-learning algorithm trains these scenarios to generate a prediction model of spall sizes even for those that have not been modeled by the physical model. Feature extraction is a key factor in the success of this approach. We extract features using two traditional approaches: statistical and physical, and an additional new approach: Time Series FeatuRe Extraction based on Scalable Hypothesis tests (TSFRESH). Experimental evaluation with well-known physical model shows that our approach achieves high accuracy, even in cases that have not been modeled by the physical model. Also, we show that the TSFRESH feature-extraction approach achieves the highest accuracy. |
| first_indexed | 2024-04-13T07:44:00Z |
| format | Article |
| id | doaj.art-9ce425576eea48c9983bb8b94f928fc3 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-13T07:44:00Z |
| publishDate | 2020-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-9ce425576eea48c9983bb8b94f928fc32022-12-22T02:55:46ZengMDPI AGSensors1424-82202020-02-01205129810.3390/s20051298s20051298Probability-Based Algorithm for Bearing Diagnosis with Untrained Spall SizesIdo Tam0Meir Kalech1Lior Rokach2Eyal Madar3Jacob Bortman4Renata Klein5Department of Software and Information Systems Engineering, Faculty of Engineering Sciences, Ben-Gurion University of the Negev, Beer-Sheva 85104, IsraelDepartment of Software and Information Systems Engineering, Faculty of Engineering Sciences, Ben-Gurion University of the Negev, Beer-Sheva 85104, IsraelDepartment of Software and Information Systems Engineering, Faculty of Engineering Sciences, Ben-Gurion University of the Negev, Beer-Sheva 85104, IsraelPearlstone Center for Aeronautical Engineering Studies and Laboratory for Mechanical Health Monitoring, Department of Mechanical Engineering, Ben-Gurion University of the Negev, P.O. Box 653, Beer Sheva 8410501, IsraelPearlstone Center for Aeronautical Engineering Studies and Laboratory for Mechanical Health Monitoring, Department of Mechanical Engineering, Ben-Gurion University of the Negev, P.O. Box 653, Beer Sheva 8410501, IsraelR.K. Diagnostics, P.O. Box 101, Gilon, D.N. Misgav 20103, IsraelBearing spall detection and predicting its size are great challenges. Model-based simulation is a well-known traditional approach to physically model the influence of the spall on the bearing. Building a physical model is challenging due to the bearing complexity and the expert knowledge required to build such a model. Obviously, building a partial physical model for some of the spall sizes is easier. In this paper, we propose a machine-learning algorithm, called Probability-Based Forest, that uses a partial physical model. First, the behavior of some of the spall sizes is physically modeled and a simulator based on this model generates scenarios for these spall sizes in different conditions. Then, the machine-learning algorithm trains these scenarios to generate a prediction model of spall sizes even for those that have not been modeled by the physical model. Feature extraction is a key factor in the success of this approach. We extract features using two traditional approaches: statistical and physical, and an additional new approach: Time Series FeatuRe Extraction based on Scalable Hypothesis tests (TSFRESH). Experimental evaluation with well-known physical model shows that our approach achieves high accuracy, even in cases that have not been modeled by the physical model. Also, we show that the TSFRESH feature-extraction approach achieves the highest accuracy.https://www.mdpi.com/1424-8220/20/5/1298machine learningbearing diagnosishybrid model |
| spellingShingle | Ido Tam Meir Kalech Lior Rokach Eyal Madar Jacob Bortman Renata Klein Probability-Based Algorithm for Bearing Diagnosis with Untrained Spall Sizes Sensors machine learning bearing diagnosis hybrid model |
| title | Probability-Based Algorithm for Bearing Diagnosis with Untrained Spall Sizes |
| title_full | Probability-Based Algorithm for Bearing Diagnosis with Untrained Spall Sizes |
| title_fullStr | Probability-Based Algorithm for Bearing Diagnosis with Untrained Spall Sizes |
| title_full_unstemmed | Probability-Based Algorithm for Bearing Diagnosis with Untrained Spall Sizes |
| title_short | Probability-Based Algorithm for Bearing Diagnosis with Untrained Spall Sizes |
| title_sort | probability based algorithm for bearing diagnosis with untrained spall sizes |
| topic | machine learning bearing diagnosis hybrid model |
| url | https://www.mdpi.com/1424-8220/20/5/1298 |
| work_keys_str_mv | AT idotam probabilitybasedalgorithmforbearingdiagnosiswithuntrainedspallsizes AT meirkalech probabilitybasedalgorithmforbearingdiagnosiswithuntrainedspallsizes AT liorrokach probabilitybasedalgorithmforbearingdiagnosiswithuntrainedspallsizes AT eyalmadar probabilitybasedalgorithmforbearingdiagnosiswithuntrainedspallsizes AT jacobbortman probabilitybasedalgorithmforbearingdiagnosiswithuntrainedspallsizes AT renataklein probabilitybasedalgorithmforbearingdiagnosiswithuntrainedspallsizes |