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

Daniel Jung, Mark Ng, Erik Frisk, Mattias Krysander

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

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.
LanguageEnglish
JournalControl Engineering Practice
Early online date9 Sep 2018
Publication statusPublished - Nov 2018

Fingerprint

Classifiers
Internal combustion engines
Failure analysis
Learning systems
Systems analysis
Sensors

Keywords

  • Fault diagnosis
  • Fault isolation
  • Machine learning
  • Artificial intelligence;
  • Classification

Cite this

@article{15980cd93c0d4546808716846b484b39,
title = "Combining model-based diagnosis and data-driven anomaly classifiers for fault isolation",
abstract = "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.",
keywords = "Fault diagnosis, Fault isolation, Machine learning, Artificial intelligence; , Classification",
author = "Daniel Jung and Mark Ng and Erik Frisk and Mattias Krysander",
year = "2018",
month = "11",
language = "English",

}

Combining model-based diagnosis and data-driven anomaly classifiers for fault isolation. / Jung, Daniel; Ng, Mark; Frisk, Erik; Krysander, Mattias.

11.2018.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Combining model-based diagnosis and data-driven anomaly classifiers for fault isolation

AU - Jung, Daniel

AU - Ng, Mark

AU - Frisk, Erik

AU - Krysander, Mattias

PY - 2018/11

Y1 - 2018/11

N2 - 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.

AB - 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.

KW - Fault diagnosis

KW - Fault isolation

KW - Machine learning

KW - Artificial intelligence;

KW - Classification

M3 - Article

ER -