Hierarchical Fault Diagnosis for a Hybrid System Based on a Multidomain Model
The diagnosis procedure is performed by integrating three steps: multidomain modeling, event identification, and failure event classification. Multidomain model can describe the normal and fault behaviors of hybrid systems efficiently and can meet the diagnosis requirements of hybrid systems. Then t...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
Hindawi Limited
2015-01-01
|
Series: | Mathematical Problems in Engineering |
Online Access: | http://dx.doi.org/10.1155/2015/361631 |
id |
doaj-76e90bd2db694b6cb6f5d53f4875ecd4 |
---|---|
record_format |
Article |
spelling |
doaj-76e90bd2db694b6cb6f5d53f4875ecd42020-11-25T00:35:13ZengHindawi LimitedMathematical Problems in Engineering1024-123X1563-51472015-01-01201510.1155/2015/361631361631Hierarchical Fault Diagnosis for a Hybrid System Based on a Multidomain ModelJiming Ma0Jianbin Guo1Sino-French Engineer School, Beihang University, No. 37 XueYuan Road, Beijing 1001091, ChinaSchool of Reliability and System Engineering, Beihang University, No. 37 XueYuan Road, Beijing 1001091, ChinaThe diagnosis procedure is performed by integrating three steps: multidomain modeling, event identification, and failure event classification. Multidomain model can describe the normal and fault behaviors of hybrid systems efficiently and can meet the diagnosis requirements of hybrid systems. Then the multidomain model is used to simulate and obtain responses under different failure events; the responses are further utilized as a priori information when training the event identification library. Finally, a brushless DC motor is selected as the study case. The experimental result indicates that the proposed method could identify the known and unknown failure events of the studied system. In particular, for a system with less response information under a failure event, the accuracy of diagnosis seems to be higher. The presented method integrates the advantages of current quantitative and qualitative diagnostic procedures and can distinguish between failures caused by parametric and abrupt structure faults. Another advantage of our method is that it can remember unknown failure types and automatically extend the adaptive resonance theory neural network library, which is extremely useful for complex hybrid systems.http://dx.doi.org/10.1155/2015/361631 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Jiming Ma Jianbin Guo |
spellingShingle |
Jiming Ma Jianbin Guo Hierarchical Fault Diagnosis for a Hybrid System Based on a Multidomain Model Mathematical Problems in Engineering |
author_facet |
Jiming Ma Jianbin Guo |
author_sort |
Jiming Ma |
title |
Hierarchical Fault Diagnosis for a Hybrid System Based on a Multidomain Model |
title_short |
Hierarchical Fault Diagnosis for a Hybrid System Based on a Multidomain Model |
title_full |
Hierarchical Fault Diagnosis for a Hybrid System Based on a Multidomain Model |
title_fullStr |
Hierarchical Fault Diagnosis for a Hybrid System Based on a Multidomain Model |
title_full_unstemmed |
Hierarchical Fault Diagnosis for a Hybrid System Based on a Multidomain Model |
title_sort |
hierarchical fault diagnosis for a hybrid system based on a multidomain model |
publisher |
Hindawi Limited |
series |
Mathematical Problems in Engineering |
issn |
1024-123X 1563-5147 |
publishDate |
2015-01-01 |
description |
The diagnosis procedure is performed by integrating three steps: multidomain modeling, event identification, and failure event classification. Multidomain model can describe the normal and fault behaviors of hybrid systems efficiently and can meet the diagnosis requirements of hybrid systems. Then the multidomain model is used to simulate and obtain responses under different failure events; the responses are further utilized as a priori information when training the event identification library. Finally, a brushless DC motor is selected as the study case. The experimental result indicates that the proposed method could identify the known and unknown failure events of the studied system. In particular, for a system with less response information under a failure event, the accuracy of diagnosis seems to be higher. The presented method integrates the advantages of current quantitative and qualitative diagnostic procedures and can distinguish between failures caused by parametric and abrupt structure faults. Another advantage of our method is that it can remember unknown failure types and automatically extend the adaptive resonance theory neural network library, which is extremely useful for complex hybrid systems. |
url |
http://dx.doi.org/10.1155/2015/361631 |
work_keys_str_mv |
AT jimingma hierarchicalfaultdiagnosisforahybridsystembasedonamultidomainmodel AT jianbinguo hierarchicalfaultdiagnosisforahybridsystembasedonamultidomainmodel |
_version_ |
1725309603508387840 |