FaultBuster: data driven fault detection and diagnosis for industrial systems

Nicola Bergantino; Fabio Caponetti; Sauro Longhi

FaultBuster: data driven fault detection and diagnosis for industrial systems

Nicola Bergantino, Fabio Caponetti, Sauro Longhi

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

423 Downloads (Pure)

Abstract

Efficient and reliable monitoring systems are mandatory to assure the required security standards in industrial complexes. This paper describes the recent developments of FaultBuster, a purely data-driven diagnostic system. It is designed so to be easily scalable to different monitor tasks. Multivariate statistical models based on principal components are used to detect abnormal situations. Tailored to alarms, a probabilistic inference engine process the fault evidences to output the most probable diagnosis. Results from the DX 09 Diagnostic Challenge shown strong detection properties, while further investigations of the diagnostic system are still needed.

Original language	English
Title of host publication	7th Workshop on Advanced Control and Diagnosis
Publication date	2009
Pages	46
Publication status	Published - 2009
Event	7th Workshop on Advanced Control and Diagnosis - Zielona Góra, Poland Duration: 19 Nov 2009 → 20 Nov 2009 Conference number: 7

Workshop

Workshop	7th Workshop on Advanced Control and Diagnosis
Number	7
Country/Territory	Poland
City	Zielona Góra
Period	19/11/2009 → 20/11/2009

Keywords

Articial intelligence
Statistical Process Control
Diagnostic inference
Fault detection

Access to Document

46 ACD 2009Final published version, 287 KB

http://www.issi.uz.zgora.pl/ACD_2009/

OpenUrl availability

Full text

Cite this

@inproceedings{968e55025efd4e5ab9a22fb8ede1946c,

title = "FaultBuster: data driven fault detection and diagnosis for industrial systems",

abstract = "Efficient and reliable monitoring systems are mandatory to assure the required security standards in industrial complexes. This paper describes the recent developments of FaultBuster, a purely data-driven diagnostic system. It is designed so to be easily scalable to different monitor tasks. Multivariate statistical models based on principal components are used to detect abnormal situations. Tailored to alarms, a probabilistic inference engine process the fault evidences to output the most probable diagnosis. Results from the DX 09 Diagnostic Challenge shown strong detection properties, while further investigations of the diagnostic system are still needed.",

keywords = "Articial intelligence, Statistical Process Control, Diagnostic inference, Fault detection",

author = "Nicola Bergantino and Fabio Caponetti and Sauro Longhi",

year = "2009",

language = "English",

pages = "46",

booktitle = "7th Workshop on Advanced Control and Diagnosis",

note = "7th Workshop on Advanced Control and Diagnosis, ACD ; Conference date: 19-11-2009 Through 20-11-2009",

}

TY - GEN

T1 - FaultBuster: data driven fault detection and diagnosis for industrial systems

AU - Bergantino, Nicola

AU - Caponetti, Fabio

AU - Longhi, Sauro

N1 - Conference code: 7

PY - 2009

Y1 - 2009

N2 - Efficient and reliable monitoring systems are mandatory to assure the required security standards in industrial complexes. This paper describes the recent developments of FaultBuster, a purely data-driven diagnostic system. It is designed so to be easily scalable to different monitor tasks. Multivariate statistical models based on principal components are used to detect abnormal situations. Tailored to alarms, a probabilistic inference engine process the fault evidences to output the most probable diagnosis. Results from the DX 09 Diagnostic Challenge shown strong detection properties, while further investigations of the diagnostic system are still needed.

AB - Efficient and reliable monitoring systems are mandatory to assure the required security standards in industrial complexes. This paper describes the recent developments of FaultBuster, a purely data-driven diagnostic system. It is designed so to be easily scalable to different monitor tasks. Multivariate statistical models based on principal components are used to detect abnormal situations. Tailored to alarms, a probabilistic inference engine process the fault evidences to output the most probable diagnosis. Results from the DX 09 Diagnostic Challenge shown strong detection properties, while further investigations of the diagnostic system are still needed.

KW - Articial intelligence

KW - Statistical Process Control

KW - Diagnostic inference

KW - Fault detection

M3 - Article in proceedings

SP - 46

BT - 7th Workshop on Advanced Control and Diagnosis

T2 - 7th Workshop on Advanced Control and Diagnosis

Y2 - 19 November 2009 through 20 November 2009

ER -

FaultBuster: data driven fault detection and diagnosis for industrial systems

Abstract

Workshop

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this