Customized neural network for sensor fusion in on-line monitoring of cutting tool wear

Choon Seong Leem, D. A. Dornfeld, S. E. Dreyfus

Research output: Contribution to journalArticle

53 Citations (Scopus)

Abstract

A customized neural network for sensor fusion of acoustic emission and force in on-line detection of tool wear is developed. Based on two critical concerns regarding practical and reliable tool-wear monitoring systems, the maximal utilization of 'unsupervised' sensor data and the avoidance of off-line feature analysis, the neural network is trained by unsupervised Kohonen's Feature Map procedure followed by an Input Feature Scaling algorithm. After levels of tool wear are topologically ordered by Kohonen's Feature Map, input features of AE and force sensor signals are transformed via Input Feature Scaling so that the resulting decision boundaries of the neural network approximate those of error-minimizing Bayes classifier. In a machining experiment, the customized neural network achieved high accuracy rates in the classification of levels of tool wear. Also, the neural network shows several practical and reliable properties for the implementation of the monitoring system in manufacturing industries.

Original languageEnglish
Pages (from-to)152-159
Number of pages8
JournalJournal of engineering for industry
Volume117
Issue number2
DOIs
Publication statusPublished - 1995 Jan 1

Fingerprint

Cutting tools
Fusion reactions
Wear of materials
Neural networks
Monitoring
Sensors
Acoustic emissions
Machining
Classifiers
Industry
Experiments

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

@article{0359894df29a4302af72de0b233437a5,
title = "Customized neural network for sensor fusion in on-line monitoring of cutting tool wear",
abstract = "A customized neural network for sensor fusion of acoustic emission and force in on-line detection of tool wear is developed. Based on two critical concerns regarding practical and reliable tool-wear monitoring systems, the maximal utilization of 'unsupervised' sensor data and the avoidance of off-line feature analysis, the neural network is trained by unsupervised Kohonen's Feature Map procedure followed by an Input Feature Scaling algorithm. After levels of tool wear are topologically ordered by Kohonen's Feature Map, input features of AE and force sensor signals are transformed via Input Feature Scaling so that the resulting decision boundaries of the neural network approximate those of error-minimizing Bayes classifier. In a machining experiment, the customized neural network achieved high accuracy rates in the classification of levels of tool wear. Also, the neural network shows several practical and reliable properties for the implementation of the monitoring system in manufacturing industries.",
author = "Leem, {Choon Seong} and Dornfeld, {D. A.} and Dreyfus, {S. E.}",
year = "1995",
month = "1",
day = "1",
doi = "10.1115/1.2803289",
language = "English",
volume = "117",
pages = "152--159",
journal = "Journal of Manufacturing Science and Engineering, Transactions of the ASME",
issn = "1087-1357",
publisher = "American Society of Mechanical Engineers(ASME)",
number = "2",

}

Customized neural network for sensor fusion in on-line monitoring of cutting tool wear. / Leem, Choon Seong; Dornfeld, D. A.; Dreyfus, S. E.

In: Journal of engineering for industry, Vol. 117, No. 2, 01.01.1995, p. 152-159.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Customized neural network for sensor fusion in on-line monitoring of cutting tool wear

AU - Leem, Choon Seong

AU - Dornfeld, D. A.

AU - Dreyfus, S. E.

PY - 1995/1/1

Y1 - 1995/1/1

N2 - A customized neural network for sensor fusion of acoustic emission and force in on-line detection of tool wear is developed. Based on two critical concerns regarding practical and reliable tool-wear monitoring systems, the maximal utilization of 'unsupervised' sensor data and the avoidance of off-line feature analysis, the neural network is trained by unsupervised Kohonen's Feature Map procedure followed by an Input Feature Scaling algorithm. After levels of tool wear are topologically ordered by Kohonen's Feature Map, input features of AE and force sensor signals are transformed via Input Feature Scaling so that the resulting decision boundaries of the neural network approximate those of error-minimizing Bayes classifier. In a machining experiment, the customized neural network achieved high accuracy rates in the classification of levels of tool wear. Also, the neural network shows several practical and reliable properties for the implementation of the monitoring system in manufacturing industries.

AB - A customized neural network for sensor fusion of acoustic emission and force in on-line detection of tool wear is developed. Based on two critical concerns regarding practical and reliable tool-wear monitoring systems, the maximal utilization of 'unsupervised' sensor data and the avoidance of off-line feature analysis, the neural network is trained by unsupervised Kohonen's Feature Map procedure followed by an Input Feature Scaling algorithm. After levels of tool wear are topologically ordered by Kohonen's Feature Map, input features of AE and force sensor signals are transformed via Input Feature Scaling so that the resulting decision boundaries of the neural network approximate those of error-minimizing Bayes classifier. In a machining experiment, the customized neural network achieved high accuracy rates in the classification of levels of tool wear. Also, the neural network shows several practical and reliable properties for the implementation of the monitoring system in manufacturing industries.

UR - http://www.scopus.com/inward/record.url?scp=0029305549&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029305549&partnerID=8YFLogxK

U2 - 10.1115/1.2803289

DO - 10.1115/1.2803289

M3 - Article

AN - SCOPUS:0029305549

VL - 117

SP - 152

EP - 159

JO - Journal of Manufacturing Science and Engineering, Transactions of the ASME

JF - Journal of Manufacturing Science and Engineering, Transactions of the ASME

SN - 1087-1357

IS - 2

ER -