Information-theoretic sensor subset selection

Application to signal-based fault isolation in diesel engines

Alok A. Joshi, Peter H. Meckl, Galen B. King, Kristofer Jennings

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper a stepwise information-theoretic feature selector is designed and implemented to reduce the dimension of a data set without losing pertinent information. The effectiveness of the proposed feature selector is demonstrated by selecting features from forty three variables monitored on a set of heavy duty diesel engines and then using this feature space for classification of faults in these engines. Using a cross-validation technique, the effects of various classification methods (linear regression, quadratic discriminants, probabilistic neural networks, and support vector machines) and feature selection methods (regression subset selection, RV-based selection by simulated annealing, and information-theoretic selection) are compared based on the percentage misclassification. The information-theoretic feature selector combined with the probabilistic neural network achieved an average classification accuracy of 90%, which was the best performance of any combination of classifiers and feature selectors under consideration.

Original languageEnglish (US)
Title of host publicationAmerican Society of Mechanical Engineers, Manufacturing Engineering Division, MED
PublisherAmerican Society of Mechanical Engineers (ASME)
ISBN (Print)0791837904, 9780791837900
DOIs
StatePublished - 2006
Externally publishedYes
Event2006 ASME International Mechanical Engineering Congress and Exposition, IMECE2006 - Chicago, IL, United States
Duration: Nov 5 2006Nov 10 2006

Other

Other2006 ASME International Mechanical Engineering Congress and Exposition, IMECE2006
CountryUnited States
CityChicago, IL
Period11/5/0611/10/06

Fingerprint

Diesel engines
Sensors
Neural networks
Simulated annealing
Set theory
Linear regression
Support vector machines
Feature extraction
Classifiers
Engines

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Joshi, A. A., Meckl, P. H., King, G. B., & Jennings, K. (2006). Information-theoretic sensor subset selection: Application to signal-based fault isolation in diesel engines. In American Society of Mechanical Engineers, Manufacturing Engineering Division, MED American Society of Mechanical Engineers (ASME). https://doi.org/10.1115/IMECE2006-15903

Information-theoretic sensor subset selection : Application to signal-based fault isolation in diesel engines. / Joshi, Alok A.; Meckl, Peter H.; King, Galen B.; Jennings, Kristofer.

American Society of Mechanical Engineers, Manufacturing Engineering Division, MED. American Society of Mechanical Engineers (ASME), 2006.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Joshi, AA, Meckl, PH, King, GB & Jennings, K 2006, Information-theoretic sensor subset selection: Application to signal-based fault isolation in diesel engines. in American Society of Mechanical Engineers, Manufacturing Engineering Division, MED. American Society of Mechanical Engineers (ASME), 2006 ASME International Mechanical Engineering Congress and Exposition, IMECE2006, Chicago, IL, United States, 11/5/06. https://doi.org/10.1115/IMECE2006-15903
Joshi AA, Meckl PH, King GB, Jennings K. Information-theoretic sensor subset selection: Application to signal-based fault isolation in diesel engines. In American Society of Mechanical Engineers, Manufacturing Engineering Division, MED. American Society of Mechanical Engineers (ASME). 2006 https://doi.org/10.1115/IMECE2006-15903
Joshi, Alok A. ; Meckl, Peter H. ; King, Galen B. ; Jennings, Kristofer. / Information-theoretic sensor subset selection : Application to signal-based fault isolation in diesel engines. American Society of Mechanical Engineers, Manufacturing Engineering Division, MED. American Society of Mechanical Engineers (ASME), 2006.
@inproceedings{ebd2154d9e96489d809b3fae5a82f09e,
title = "Information-theoretic sensor subset selection: Application to signal-based fault isolation in diesel engines",
abstract = "In this paper a stepwise information-theoretic feature selector is designed and implemented to reduce the dimension of a data set without losing pertinent information. The effectiveness of the proposed feature selector is demonstrated by selecting features from forty three variables monitored on a set of heavy duty diesel engines and then using this feature space for classification of faults in these engines. Using a cross-validation technique, the effects of various classification methods (linear regression, quadratic discriminants, probabilistic neural networks, and support vector machines) and feature selection methods (regression subset selection, RV-based selection by simulated annealing, and information-theoretic selection) are compared based on the percentage misclassification. The information-theoretic feature selector combined with the probabilistic neural network achieved an average classification accuracy of 90{\%}, which was the best performance of any combination of classifiers and feature selectors under consideration.",
author = "Joshi, {Alok A.} and Meckl, {Peter H.} and King, {Galen B.} and Kristofer Jennings",
year = "2006",
doi = "10.1115/IMECE2006-15903",
language = "English (US)",
isbn = "0791837904",
booktitle = "American Society of Mechanical Engineers, Manufacturing Engineering Division, MED",
publisher = "American Society of Mechanical Engineers (ASME)",

}

TY - GEN

T1 - Information-theoretic sensor subset selection

T2 - Application to signal-based fault isolation in diesel engines

AU - Joshi, Alok A.

AU - Meckl, Peter H.

AU - King, Galen B.

AU - Jennings, Kristofer

PY - 2006

Y1 - 2006

N2 - In this paper a stepwise information-theoretic feature selector is designed and implemented to reduce the dimension of a data set without losing pertinent information. The effectiveness of the proposed feature selector is demonstrated by selecting features from forty three variables monitored on a set of heavy duty diesel engines and then using this feature space for classification of faults in these engines. Using a cross-validation technique, the effects of various classification methods (linear regression, quadratic discriminants, probabilistic neural networks, and support vector machines) and feature selection methods (regression subset selection, RV-based selection by simulated annealing, and information-theoretic selection) are compared based on the percentage misclassification. The information-theoretic feature selector combined with the probabilistic neural network achieved an average classification accuracy of 90%, which was the best performance of any combination of classifiers and feature selectors under consideration.

AB - In this paper a stepwise information-theoretic feature selector is designed and implemented to reduce the dimension of a data set without losing pertinent information. The effectiveness of the proposed feature selector is demonstrated by selecting features from forty three variables monitored on a set of heavy duty diesel engines and then using this feature space for classification of faults in these engines. Using a cross-validation technique, the effects of various classification methods (linear regression, quadratic discriminants, probabilistic neural networks, and support vector machines) and feature selection methods (regression subset selection, RV-based selection by simulated annealing, and information-theoretic selection) are compared based on the percentage misclassification. The information-theoretic feature selector combined with the probabilistic neural network achieved an average classification accuracy of 90%, which was the best performance of any combination of classifiers and feature selectors under consideration.

UR - http://www.scopus.com/inward/record.url?scp=84920634147&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84920634147&partnerID=8YFLogxK

U2 - 10.1115/IMECE2006-15903

DO - 10.1115/IMECE2006-15903

M3 - Conference contribution

SN - 0791837904

SN - 9780791837900

BT - American Society of Mechanical Engineers, Manufacturing Engineering Division, MED

PB - American Society of Mechanical Engineers (ASME)

ER -