Hybrid reasoning for prognostic learning in CBM systems

Amulya K. Garga, Katherine T. McClintic, Robert Lee Campbell, Chih Chung Yang, Mitchell S. Lebold, Todd A. Hay, Carl S. Byington

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

Reasoning systems that integrate explicit knowledge with implicit information are essential for high performance decision support in condition-based maintenance and prognostic health management applications. Such reasoning systems must be capable of learning the specific features of each machine during its life cycle. In this paper, a hybrid reasoning approach that is capable of integrating domain knowledge and test and operational data from the machine is described. This approach is illustrated with an industrial gearbox example. In this approach explicit domain knowledge is expressed as a rule-base and used to train a feed-forward neural network. The training process results in a parsimonious representation of the explicit knowledge by combining redundant rules. A significant added practical benefit of this process is that it also is able to identify logical inconsistencies in the rule-base. Such inconsistencies are notorious in causing deadlock in large-scale expert systems. The neural network can be periodically updated with test and operational data to adapt the network to each specific machine. The flexibility and efficiency of this hybrid approach make it very suitable for practical health management systems designed to operate in a distributed environment.

Original languageEnglish (US)
Pages (from-to)62957-62969
Number of pages13
JournalIEEE Aerospace Conference Proceedings
Volume6
DOIs
StatePublished - Jan 1 2001

Fingerprint

learning
Health
Feedforward neural networks
Expert systems
health
Life cycle
Neural networks
transmissions (machine elements)
expert systems
management systems
expert system
maintenance
train
flexibility
education
life cycle
cycles
test

All Science Journal Classification (ASJC) codes

  • Aerospace Engineering
  • Space and Planetary Science

Cite this

Garga, A. K., McClintic, K. T., Campbell, R. L., Yang, C. C., Lebold, M. S., Hay, T. A., & Byington, C. S. (2001). Hybrid reasoning for prognostic learning in CBM systems. IEEE Aerospace Conference Proceedings, 6, 62957-62969. https://doi.org/10.1109/AERO.2001.931316
Garga, Amulya K. ; McClintic, Katherine T. ; Campbell, Robert Lee ; Yang, Chih Chung ; Lebold, Mitchell S. ; Hay, Todd A. ; Byington, Carl S. / Hybrid reasoning for prognostic learning in CBM systems. In: IEEE Aerospace Conference Proceedings. 2001 ; Vol. 6. pp. 62957-62969.
@article{82ce208fa58d424a982766204e989613,
title = "Hybrid reasoning for prognostic learning in CBM systems",
abstract = "Reasoning systems that integrate explicit knowledge with implicit information are essential for high performance decision support in condition-based maintenance and prognostic health management applications. Such reasoning systems must be capable of learning the specific features of each machine during its life cycle. In this paper, a hybrid reasoning approach that is capable of integrating domain knowledge and test and operational data from the machine is described. This approach is illustrated with an industrial gearbox example. In this approach explicit domain knowledge is expressed as a rule-base and used to train a feed-forward neural network. The training process results in a parsimonious representation of the explicit knowledge by combining redundant rules. A significant added practical benefit of this process is that it also is able to identify logical inconsistencies in the rule-base. Such inconsistencies are notorious in causing deadlock in large-scale expert systems. The neural network can be periodically updated with test and operational data to adapt the network to each specific machine. The flexibility and efficiency of this hybrid approach make it very suitable for practical health management systems designed to operate in a distributed environment.",
author = "Garga, {Amulya K.} and McClintic, {Katherine T.} and Campbell, {Robert Lee} and Yang, {Chih Chung} and Lebold, {Mitchell S.} and Hay, {Todd A.} and Byington, {Carl S.}",
year = "2001",
month = "1",
day = "1",
doi = "10.1109/AERO.2001.931316",
language = "English (US)",
volume = "6",
pages = "62957--62969",
journal = "IEEE Aerospace Conference Proceedings",
issn = "1095-323X",
publisher = "IEEE Computer Society",

}

Garga, AK, McClintic, KT, Campbell, RL, Yang, CC, Lebold, MS, Hay, TA & Byington, CS 2001, 'Hybrid reasoning for prognostic learning in CBM systems', IEEE Aerospace Conference Proceedings, vol. 6, pp. 62957-62969. https://doi.org/10.1109/AERO.2001.931316

Hybrid reasoning for prognostic learning in CBM systems. / Garga, Amulya K.; McClintic, Katherine T.; Campbell, Robert Lee; Yang, Chih Chung; Lebold, Mitchell S.; Hay, Todd A.; Byington, Carl S.

In: IEEE Aerospace Conference Proceedings, Vol. 6, 01.01.2001, p. 62957-62969.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Hybrid reasoning for prognostic learning in CBM systems

AU - Garga, Amulya K.

AU - McClintic, Katherine T.

AU - Campbell, Robert Lee

AU - Yang, Chih Chung

AU - Lebold, Mitchell S.

AU - Hay, Todd A.

AU - Byington, Carl S.

PY - 2001/1/1

Y1 - 2001/1/1

N2 - Reasoning systems that integrate explicit knowledge with implicit information are essential for high performance decision support in condition-based maintenance and prognostic health management applications. Such reasoning systems must be capable of learning the specific features of each machine during its life cycle. In this paper, a hybrid reasoning approach that is capable of integrating domain knowledge and test and operational data from the machine is described. This approach is illustrated with an industrial gearbox example. In this approach explicit domain knowledge is expressed as a rule-base and used to train a feed-forward neural network. The training process results in a parsimonious representation of the explicit knowledge by combining redundant rules. A significant added practical benefit of this process is that it also is able to identify logical inconsistencies in the rule-base. Such inconsistencies are notorious in causing deadlock in large-scale expert systems. The neural network can be periodically updated with test and operational data to adapt the network to each specific machine. The flexibility and efficiency of this hybrid approach make it very suitable for practical health management systems designed to operate in a distributed environment.

AB - Reasoning systems that integrate explicit knowledge with implicit information are essential for high performance decision support in condition-based maintenance and prognostic health management applications. Such reasoning systems must be capable of learning the specific features of each machine during its life cycle. In this paper, a hybrid reasoning approach that is capable of integrating domain knowledge and test and operational data from the machine is described. This approach is illustrated with an industrial gearbox example. In this approach explicit domain knowledge is expressed as a rule-base and used to train a feed-forward neural network. The training process results in a parsimonious representation of the explicit knowledge by combining redundant rules. A significant added practical benefit of this process is that it also is able to identify logical inconsistencies in the rule-base. Such inconsistencies are notorious in causing deadlock in large-scale expert systems. The neural network can be periodically updated with test and operational data to adapt the network to each specific machine. The flexibility and efficiency of this hybrid approach make it very suitable for practical health management systems designed to operate in a distributed environment.

UR - http://www.scopus.com/inward/record.url?scp=0034856633&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034856633&partnerID=8YFLogxK

U2 - 10.1109/AERO.2001.931316

DO - 10.1109/AERO.2001.931316

M3 - Article

VL - 6

SP - 62957

EP - 62969

JO - IEEE Aerospace Conference Proceedings

JF - IEEE Aerospace Conference Proceedings

SN - 1095-323X

ER -