Learning input feature selection for sensor fusion in tool wear monitoring

Choon Seong Leem, Stuart E. Dreyfus

Research output: Contribution to conferencePaper

8 Citations (Scopus)

Abstract

We develop a neural network for on-line tool wear monitoring in metal cutting environments. After levels of tool wear are topologically ordered by the unsupervised Kohonen's Feature Map, input features from Acoustic Emission and force sensor signals are scaled by an additional supervised learning stage using Input Feature Scaling(IFS) algorithm developed in this work. In a machining experiment, without any off-line feature selection procedure, this neural network with the ability to learn feature selection achieves 94% and 92% accuracy for classification into two and three levels of tool wear, respectively. In conjunction with Kohonen's Feature Map, IFS is a practical and reliable pattern classifier for sensor fusion.

Original languageEnglish
Pages815-820
Number of pages6
Publication statusPublished - 1992 Dec 1
EventProceedings of the 1992 Artificial Neural Networks in Engineering, ANNIE'92 - St.Louis, MO, USA
Duration: 1992 Nov 151992 Nov 18

Other

OtherProceedings of the 1992 Artificial Neural Networks in Engineering, ANNIE'92
CitySt.Louis, MO, USA
Period92/11/1592/11/18

Fingerprint

Feature extraction
Fusion reactions
Wear of materials
Monitoring
Sensors
Neural networks
Metal cutting
Supervised learning
Acoustic emissions
Machining
Classifiers
Experiments

All Science Journal Classification (ASJC) codes

  • Software

Cite this

Leem, C. S., & Dreyfus, S. E. (1992). Learning input feature selection for sensor fusion in tool wear monitoring. 815-820. Paper presented at Proceedings of the 1992 Artificial Neural Networks in Engineering, ANNIE'92, St.Louis, MO, USA, .
Leem, Choon Seong ; Dreyfus, Stuart E. / Learning input feature selection for sensor fusion in tool wear monitoring. Paper presented at Proceedings of the 1992 Artificial Neural Networks in Engineering, ANNIE'92, St.Louis, MO, USA, .6 p.
@conference{cf5dc1a2eb464e8eb016afbb82e55263,
title = "Learning input feature selection for sensor fusion in tool wear monitoring",
abstract = "We develop a neural network for on-line tool wear monitoring in metal cutting environments. After levels of tool wear are topologically ordered by the unsupervised Kohonen's Feature Map, input features from Acoustic Emission and force sensor signals are scaled by an additional supervised learning stage using Input Feature Scaling(IFS) algorithm developed in this work. In a machining experiment, without any off-line feature selection procedure, this neural network with the ability to learn feature selection achieves 94{\%} and 92{\%} accuracy for classification into two and three levels of tool wear, respectively. In conjunction with Kohonen's Feature Map, IFS is a practical and reliable pattern classifier for sensor fusion.",
author = "Leem, {Choon Seong} and Dreyfus, {Stuart E.}",
year = "1992",
month = "12",
day = "1",
language = "English",
pages = "815--820",
note = "Proceedings of the 1992 Artificial Neural Networks in Engineering, ANNIE'92 ; Conference date: 15-11-1992 Through 18-11-1992",

}

Leem, CS & Dreyfus, SE 1992, 'Learning input feature selection for sensor fusion in tool wear monitoring', Paper presented at Proceedings of the 1992 Artificial Neural Networks in Engineering, ANNIE'92, St.Louis, MO, USA, 92/11/15 - 92/11/18 pp. 815-820.

Learning input feature selection for sensor fusion in tool wear monitoring. / Leem, Choon Seong; Dreyfus, Stuart E.

1992. 815-820 Paper presented at Proceedings of the 1992 Artificial Neural Networks in Engineering, ANNIE'92, St.Louis, MO, USA, .

Research output: Contribution to conferencePaper

TY - CONF

T1 - Learning input feature selection for sensor fusion in tool wear monitoring

AU - Leem, Choon Seong

AU - Dreyfus, Stuart E.

PY - 1992/12/1

Y1 - 1992/12/1

N2 - We develop a neural network for on-line tool wear monitoring in metal cutting environments. After levels of tool wear are topologically ordered by the unsupervised Kohonen's Feature Map, input features from Acoustic Emission and force sensor signals are scaled by an additional supervised learning stage using Input Feature Scaling(IFS) algorithm developed in this work. In a machining experiment, without any off-line feature selection procedure, this neural network with the ability to learn feature selection achieves 94% and 92% accuracy for classification into two and three levels of tool wear, respectively. In conjunction with Kohonen's Feature Map, IFS is a practical and reliable pattern classifier for sensor fusion.

AB - We develop a neural network for on-line tool wear monitoring in metal cutting environments. After levels of tool wear are topologically ordered by the unsupervised Kohonen's Feature Map, input features from Acoustic Emission and force sensor signals are scaled by an additional supervised learning stage using Input Feature Scaling(IFS) algorithm developed in this work. In a machining experiment, without any off-line feature selection procedure, this neural network with the ability to learn feature selection achieves 94% and 92% accuracy for classification into two and three levels of tool wear, respectively. In conjunction with Kohonen's Feature Map, IFS is a practical and reliable pattern classifier for sensor fusion.

UR - http://www.scopus.com/inward/record.url?scp=0027105437&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027105437&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0027105437

SP - 815

EP - 820

ER -

Leem CS, Dreyfus SE. Learning input feature selection for sensor fusion in tool wear monitoring. 1992. Paper presented at Proceedings of the 1992 Artificial Neural Networks in Engineering, ANNIE'92, St.Louis, MO, USA, .