A combined diagnosis system design using model-based and data-driven methods

Daniel Jung; Kok Ng Ng; Erik Frisk; Mattias Krysander

doi:10.1109/SYSTOL.2016.7739747

A combined diagnosis system design using model-based and data-driven methods

Daniel Jung, Kok Ng Ng, Erik Frisk, Mattias Krysander

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

12 Citations (Scopus)

Abstract

A hybrid diagnosis system design is proposed that combines model-based and data-driven diagnosis methods for fault isolation. A set of residuals are used to detect if there is a fault in the system and a consistency-based fault isolation algorithm is used to compute all diagnosis candidates that can explain the triggered residuals. To improve fault isolation, diagnosis candidates are ranked by evaluating the residuals using a set of one-class support vector machines trained using data from different faults. The proposed diagnosis system design is evaluated using simulations of a model describing the air-flow in an internal combustion engine.

Original language	English
Title of host publication	Unknown Host Publication
Publisher	IEEE
Pages	177-182
Number of pages	6
ISBN (Electronic)	978-1-5090-0658-8
ISBN (Print)	9781-5090-0657-1, 978-1-5090-0659-5
DOIs	https://doi.org/10.1109/SYSTOL.2016.7739747
Publication status	Published - 10 Nov 2016
Event	2016 3rd Conference on Control and Fault-Tolerant Systems (SysTol) Barcelona, Spain - Barcelona, Spain Duration: 17 Jun 2016 → …

Conference

Conference	2016 3rd Conference on Control and Fault-Tolerant Systems (SysTol) Barcelona, Spain
Period	17/06/16 → …

Bibliographical note

Not compliant in UIR, but author was not employed by UK HEI at time of submission, so exception set (see 'Other files' for evidence)

Keywords

fault diagnosis
model-based
data-driven
vehicular system

Access to Document

10.1109/SYSTOL.2016.7739747

http://uir.ulster.ac.uk/38284/3/SysTol_16_DJ_MN_EF_MK.pdf

Cite this

@inproceedings{75f5f5aef82d4e788aedddf15bee60e3,

title = "A combined diagnosis system design using model-based and data-driven methods",

abstract = "A hybrid diagnosis system design is proposed that combines model-based and data-driven diagnosis methods for fault isolation. A set of residuals are used to detect if there is a fault in the system and a consistency-based fault isolation algorithm is used to compute all diagnosis candidates that can explain the triggered residuals. To improve fault isolation, diagnosis candidates are ranked by evaluating the residuals using a set of one-class support vector machines trained using data from different faults. The proposed diagnosis system design is evaluated using simulations of a model describing the air-flow in an internal combustion engine.",

keywords = "fault diagnosis, model-based, data-driven, vehicular system",

author = "Daniel Jung and Ng, {Kok Ng} and Erik Frisk and Mattias Krysander",

note = "Not compliant in UIR, but author was not employed by UK HEI at time of submission, so exception set (see 'Other files' for evidence) ; 2016 3rd Conference on Control and Fault-Tolerant Systems (SysTol) Barcelona, Spain ; Conference date: 17-06-2016",

year = "2016",

month = nov,

day = "10",

doi = "10.1109/SYSTOL.2016.7739747",

language = "English",

isbn = "9781-5090-0657-1",

pages = "177--182",

booktitle = "Unknown Host Publication",

publisher = "IEEE",

address = "United States",

}

TY - GEN

T1 - A combined diagnosis system design using model-based and data-driven methods

AU - Jung, Daniel

AU - Ng, Kok Ng

AU - Frisk, Erik

AU - Krysander, Mattias

N1 - Not compliant in UIR, but author was not employed by UK HEI at time of submission, so exception set (see 'Other files' for evidence)

PY - 2016/11/10

Y1 - 2016/11/10

N2 - A hybrid diagnosis system design is proposed that combines model-based and data-driven diagnosis methods for fault isolation. A set of residuals are used to detect if there is a fault in the system and a consistency-based fault isolation algorithm is used to compute all diagnosis candidates that can explain the triggered residuals. To improve fault isolation, diagnosis candidates are ranked by evaluating the residuals using a set of one-class support vector machines trained using data from different faults. The proposed diagnosis system design is evaluated using simulations of a model describing the air-flow in an internal combustion engine.

AB - A hybrid diagnosis system design is proposed that combines model-based and data-driven diagnosis methods for fault isolation. A set of residuals are used to detect if there is a fault in the system and a consistency-based fault isolation algorithm is used to compute all diagnosis candidates that can explain the triggered residuals. To improve fault isolation, diagnosis candidates are ranked by evaluating the residuals using a set of one-class support vector machines trained using data from different faults. The proposed diagnosis system design is evaluated using simulations of a model describing the air-flow in an internal combustion engine.

KW - fault diagnosis

KW - model-based

KW - data-driven

KW - vehicular system

UR - https://pure.ulster.ac.uk/en/publications/a-combined-diagnosis-system-design-using-model-based-and-data-dri

U2 - 10.1109/SYSTOL.2016.7739747

DO - 10.1109/SYSTOL.2016.7739747

M3 - Conference contribution

SN - 9781-5090-0657-1

SN - 978-1-5090-0659-5

SP - 177

EP - 182

BT - Unknown Host Publication

PB - IEEE

T2 - 2016 3rd Conference on Control and Fault-Tolerant Systems (SysTol) Barcelona, Spain

Y2 - 17 June 2016

ER -

A combined diagnosis system design using model-based and data-driven methods

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Design and Selection of Additional Residuals to Enhance Fault Isolation of a Turbocharged Spark Ignited Engine System

Cite this