Combining model-based diagnosis and data-driven anomaly classifiers for fault isolation

Machine learning can be used to automatically process sensor data and create data-driven models for prediction and classification. However, in applications such as fault diagnosis, faults are rare events and learning models for fault classification is complicated because of lack of relevant training...

Full description

Saved in:
Bibliographic Details
Published inControl engineering practice Vol. 80; pp. 146 - 156
Main Authors Jung, Daniel, Ng, Kok Yew, Frisk, Erik, Krysander, Mattias
Format Journal Article
LanguageEnglish
Published Elsevier Ltd 01.11.2018
Subjects
Artificial intelligence
Classification
Fault diagnosis
Fault isolation
Machine learning
Fault diagnosis
Fault isolation
Artificial intelligence
Machine learning
Classification
Online AccessGet full text
ISSN0967-0661
1873-6939
1873-6939
DOI10.1016/j.conengprac.2018.08.013

Cover

More Information
Summary:Machine learning can be used to automatically process sensor data and create data-driven models for prediction and classification. However, in applications such as fault diagnosis, faults are rare events and learning models for fault classification is complicated because of lack of relevant training data. This paper proposes a hybrid diagnosis system design which combines model-based residuals with incremental anomaly classifiers. The proposed method is able to identify unknown faults and also classify multiple-faults using only single-fault training data. The proposed method is verified using a physical model and data collected from an internal combustion engine.
ISSN:0967-0661
1873-6939
1873-6939
DOI:10.1016/j.conengprac.2018.08.013