The Robust Multi-Scale Deep-SVDD Model for Anomaly Online Detection of Rolling Bearings
Aiming at the online detection problem of rolling bearings, the limited amount of target bearing data leads to insufficient model in training and feature representation. It is difficult for the online detection model to construct an accurate decision boundary. To solve the problem, a multi-scale rob...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-07-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/22/15/5681 |
_version_ | 1797412319424151552 |
---|---|
author | Linlin Kou Jiaxian Chen Yong Qin Wentao Mao |
author_facet | Linlin Kou Jiaxian Chen Yong Qin Wentao Mao |
author_sort | Linlin Kou |
collection | DOAJ |
description | Aiming at the online detection problem of rolling bearings, the limited amount of target bearing data leads to insufficient model in training and feature representation. It is difficult for the online detection model to construct an accurate decision boundary. To solve the problem, a multi-scale robust anomaly detection method based on data enhancement technology is proposed in this paper. Firstly, the training data are transformed into multiple subspaces through the data enhancement technology. Then, a prototype clustering method is introduced to enhance the robustness of features representation under the framework of the robust deep auto-encoding algorithm. Finally, the robust multi-scale Deep-SVDD hyper sphere model is constructed to achieve online detection of abnormal state data. Experiments are conducted on the IEEE PHM Challenge 2012 bearing data set and XJTU-TU data set. The proposed method shows much greater susceptibility to incipient faults, and it has fewer false alarms. The robust multi-scale Deep-SVDD hyper sphere model significantly improves the performance of incipient fault detection for rolling bearings. |
first_indexed | 2024-03-09T05:00:27Z |
format | Article |
id | doaj.art-c84c6c9db1194de9a34e0142549d2c25 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-09T05:00:27Z |
publishDate | 2022-07-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-c84c6c9db1194de9a34e0142549d2c252023-12-03T13:01:00ZengMDPI AGSensors1424-82202022-07-012215568110.3390/s22155681The Robust Multi-Scale Deep-SVDD Model for Anomaly Online Detection of Rolling BearingsLinlin Kou0Jiaxian Chen1Yong Qin2Wentao Mao3State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing 100044, ChinaCollege of Computer and Information Engineering, Henan Normal University, Xinxiang 453007, ChinaState Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing 100044, ChinaCollege of Computer and Information Engineering, Henan Normal University, Xinxiang 453007, ChinaAiming at the online detection problem of rolling bearings, the limited amount of target bearing data leads to insufficient model in training and feature representation. It is difficult for the online detection model to construct an accurate decision boundary. To solve the problem, a multi-scale robust anomaly detection method based on data enhancement technology is proposed in this paper. Firstly, the training data are transformed into multiple subspaces through the data enhancement technology. Then, a prototype clustering method is introduced to enhance the robustness of features representation under the framework of the robust deep auto-encoding algorithm. Finally, the robust multi-scale Deep-SVDD hyper sphere model is constructed to achieve online detection of abnormal state data. Experiments are conducted on the IEEE PHM Challenge 2012 bearing data set and XJTU-TU data set. The proposed method shows much greater susceptibility to incipient faults, and it has fewer false alarms. The robust multi-scale Deep-SVDD hyper sphere model significantly improves the performance of incipient fault detection for rolling bearings.https://www.mdpi.com/1424-8220/22/15/5681incipient fault detectionrobustnessreinforcement learninganomaly detection |
spellingShingle | Linlin Kou Jiaxian Chen Yong Qin Wentao Mao The Robust Multi-Scale Deep-SVDD Model for Anomaly Online Detection of Rolling Bearings Sensors incipient fault detection robustness reinforcement learning anomaly detection |
title | The Robust Multi-Scale Deep-SVDD Model for Anomaly Online Detection of Rolling Bearings |
title_full | The Robust Multi-Scale Deep-SVDD Model for Anomaly Online Detection of Rolling Bearings |
title_fullStr | The Robust Multi-Scale Deep-SVDD Model for Anomaly Online Detection of Rolling Bearings |
title_full_unstemmed | The Robust Multi-Scale Deep-SVDD Model for Anomaly Online Detection of Rolling Bearings |
title_short | The Robust Multi-Scale Deep-SVDD Model for Anomaly Online Detection of Rolling Bearings |
title_sort | robust multi scale deep svdd model for anomaly online detection of rolling bearings |
topic | incipient fault detection robustness reinforcement learning anomaly detection |
url | https://www.mdpi.com/1424-8220/22/15/5681 |
work_keys_str_mv | AT linlinkou therobustmultiscaledeepsvddmodelforanomalyonlinedetectionofrollingbearings AT jiaxianchen therobustmultiscaledeepsvddmodelforanomalyonlinedetectionofrollingbearings AT yongqin therobustmultiscaledeepsvddmodelforanomalyonlinedetectionofrollingbearings AT wentaomao therobustmultiscaledeepsvddmodelforanomalyonlinedetectionofrollingbearings AT linlinkou robustmultiscaledeepsvddmodelforanomalyonlinedetectionofrollingbearings AT jiaxianchen robustmultiscaledeepsvddmodelforanomalyonlinedetectionofrollingbearings AT yongqin robustmultiscaledeepsvddmodelforanomalyonlinedetectionofrollingbearings AT wentaomao robustmultiscaledeepsvddmodelforanomalyonlinedetectionofrollingbearings |