Performance comparison of machine learning algorithms and feature reduction techniques for tool condition classification during end milling

Sultan Binsaeid, Shihab S Asfour, Sohyung Cho, Moataz Mohamed Eltoukhy

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

Abstract

In fully automated machining environments, the immediate detection of condition state of the cutting tool that is identified as worn, chipped, or broken tool is essential to the improvement of productivity and cost effectiveness. In this paper, a multi-sensor tool condition monitoring system (TCM) via machine learning (ML) approach is proposed to identify the instantaneous condition of a multi-layer coated and multi-flute carbide end mill cutter when machining 4340 steel based on extracted sensors signal features. Using data acquisition and signal processing for force, vibration, acoustic emission, and spindle power sensor, 135 different features were extracted from all sensory signals in the time and frequency domain. Then, these features along with machining parameters are evaluated for significance using different feature reduction techniques. Two feature extraction methods were investigated; independent component analysis (ICA), and principle component analysis (PCA) and three feature selection methods were studied; Chi square, Information Gain, and ReliefF. Five classifiers were used to assess the effectiveness of the different feature reduction techniques. These classifiers were Naïve Bayes Tree, C4.5, Bayesian Network, Nearest Neighbor, and K-star algorithms. These selected ML algorithms classify not only flank wear level of the multi-flute end mill but also breakage and chipping. The criterion for selecting the best model depends on classification accuracy and simplicity of the implemented system where the number of features and sensors are kept to minimum in order to increase the efficiency of the online acquisition system. In this paper, all feature selection methods have relatively outperformed the extraction methods using ICA and PCA. In addition, the utilization of simple classifiers such as Nearest Neighbor and K-star classifier has proven to be effective in classifying tools abnormalities in end milling.

Original languageEnglish
Title of host publication37th International Conference on Computers and Industrial Engineering 2007
Pages2378-2389
Number of pages12
Volume3
StatePublished - Dec 1 2007
Event37th International Conference on Computers and Industrial Engineering 2007 - Alexandria, Egypt
Duration: Oct 20 2007Oct 23 2007

Other

Other37th International Conference on Computers and Industrial Engineering 2007
CountryEgypt
CityAlexandria
Period10/20/0710/23/07

Fingerprint

Milling (machining)
Learning algorithms
Learning systems
Classifiers
Feature extraction
Machining
Sensors
Independent component analysis
Stars
Condition monitoring
Bayesian networks
Cutting tools
Cost effectiveness
Acoustic emissions
Vibrations (mechanical)
Carbides
Data acquisition
Signal processing
Productivity
Wear of materials

Keywords

  • Feature selection
  • Machine Learning
  • Tool condition monitoring
  • Tool wear

ASJC Scopus subject areas

  • Computer Science (miscellaneous)
  • Industrial and Manufacturing Engineering

Cite this

Binsaeid, S., Asfour, S. S., Cho, S., & Eltoukhy, M. M. (2007). Performance comparison of machine learning algorithms and feature reduction techniques for tool condition classification during end milling. In 37th International Conference on Computers and Industrial Engineering 2007 (Vol. 3, pp. 2378-2389)

Performance comparison of machine learning algorithms and feature reduction techniques for tool condition classification during end milling. / Binsaeid, Sultan; Asfour, Shihab S; Cho, Sohyung; Eltoukhy, Moataz Mohamed.

37th International Conference on Computers and Industrial Engineering 2007. Vol. 3 2007. p. 2378-2389.

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

Binsaeid, S, Asfour, SS, Cho, S & Eltoukhy, MM 2007, Performance comparison of machine learning algorithms and feature reduction techniques for tool condition classification during end milling. in 37th International Conference on Computers and Industrial Engineering 2007. vol. 3, pp. 2378-2389, 37th International Conference on Computers and Industrial Engineering 2007, Alexandria, Egypt, 10/20/07.
Binsaeid S, Asfour SS, Cho S, Eltoukhy MM. Performance comparison of machine learning algorithms and feature reduction techniques for tool condition classification during end milling. In 37th International Conference on Computers and Industrial Engineering 2007. Vol. 3. 2007. p. 2378-2389
Binsaeid, Sultan ; Asfour, Shihab S ; Cho, Sohyung ; Eltoukhy, Moataz Mohamed. / Performance comparison of machine learning algorithms and feature reduction techniques for tool condition classification during end milling. 37th International Conference on Computers and Industrial Engineering 2007. Vol. 3 2007. pp. 2378-2389
@inproceedings{0ffa8b971d55405c844dc0fc192ca66f,
title = "Performance comparison of machine learning algorithms and feature reduction techniques for tool condition classification during end milling",
abstract = "In fully automated machining environments, the immediate detection of condition state of the cutting tool that is identified as worn, chipped, or broken tool is essential to the improvement of productivity and cost effectiveness. In this paper, a multi-sensor tool condition monitoring system (TCM) via machine learning (ML) approach is proposed to identify the instantaneous condition of a multi-layer coated and multi-flute carbide end mill cutter when machining 4340 steel based on extracted sensors signal features. Using data acquisition and signal processing for force, vibration, acoustic emission, and spindle power sensor, 135 different features were extracted from all sensory signals in the time and frequency domain. Then, these features along with machining parameters are evaluated for significance using different feature reduction techniques. Two feature extraction methods were investigated; independent component analysis (ICA), and principle component analysis (PCA) and three feature selection methods were studied; Chi square, Information Gain, and ReliefF. Five classifiers were used to assess the effectiveness of the different feature reduction techniques. These classifiers were Na{\"i}ve Bayes Tree, C4.5, Bayesian Network, Nearest Neighbor, and K-star algorithms. These selected ML algorithms classify not only flank wear level of the multi-flute end mill but also breakage and chipping. The criterion for selecting the best model depends on classification accuracy and simplicity of the implemented system where the number of features and sensors are kept to minimum in order to increase the efficiency of the online acquisition system. In this paper, all feature selection methods have relatively outperformed the extraction methods using ICA and PCA. In addition, the utilization of simple classifiers such as Nearest Neighbor and K-star classifier has proven to be effective in classifying tools abnormalities in end milling.",
keywords = "Feature selection, Machine Learning, Tool condition monitoring, Tool wear",
author = "Sultan Binsaeid and Asfour, {Shihab S} and Sohyung Cho and Eltoukhy, {Moataz Mohamed}",
year = "2007",
month = "12",
day = "1",
language = "English",
isbn = "9781627486811",
volume = "3",
pages = "2378--2389",
booktitle = "37th International Conference on Computers and Industrial Engineering 2007",

}

TY - GEN

T1 - Performance comparison of machine learning algorithms and feature reduction techniques for tool condition classification during end milling

AU - Binsaeid, Sultan

AU - Asfour, Shihab S

AU - Cho, Sohyung

AU - Eltoukhy, Moataz Mohamed

PY - 2007/12/1

Y1 - 2007/12/1

N2 - In fully automated machining environments, the immediate detection of condition state of the cutting tool that is identified as worn, chipped, or broken tool is essential to the improvement of productivity and cost effectiveness. In this paper, a multi-sensor tool condition monitoring system (TCM) via machine learning (ML) approach is proposed to identify the instantaneous condition of a multi-layer coated and multi-flute carbide end mill cutter when machining 4340 steel based on extracted sensors signal features. Using data acquisition and signal processing for force, vibration, acoustic emission, and spindle power sensor, 135 different features were extracted from all sensory signals in the time and frequency domain. Then, these features along with machining parameters are evaluated for significance using different feature reduction techniques. Two feature extraction methods were investigated; independent component analysis (ICA), and principle component analysis (PCA) and three feature selection methods were studied; Chi square, Information Gain, and ReliefF. Five classifiers were used to assess the effectiveness of the different feature reduction techniques. These classifiers were Naïve Bayes Tree, C4.5, Bayesian Network, Nearest Neighbor, and K-star algorithms. These selected ML algorithms classify not only flank wear level of the multi-flute end mill but also breakage and chipping. The criterion for selecting the best model depends on classification accuracy and simplicity of the implemented system where the number of features and sensors are kept to minimum in order to increase the efficiency of the online acquisition system. In this paper, all feature selection methods have relatively outperformed the extraction methods using ICA and PCA. In addition, the utilization of simple classifiers such as Nearest Neighbor and K-star classifier has proven to be effective in classifying tools abnormalities in end milling.

AB - In fully automated machining environments, the immediate detection of condition state of the cutting tool that is identified as worn, chipped, or broken tool is essential to the improvement of productivity and cost effectiveness. In this paper, a multi-sensor tool condition monitoring system (TCM) via machine learning (ML) approach is proposed to identify the instantaneous condition of a multi-layer coated and multi-flute carbide end mill cutter when machining 4340 steel based on extracted sensors signal features. Using data acquisition and signal processing for force, vibration, acoustic emission, and spindle power sensor, 135 different features were extracted from all sensory signals in the time and frequency domain. Then, these features along with machining parameters are evaluated for significance using different feature reduction techniques. Two feature extraction methods were investigated; independent component analysis (ICA), and principle component analysis (PCA) and three feature selection methods were studied; Chi square, Information Gain, and ReliefF. Five classifiers were used to assess the effectiveness of the different feature reduction techniques. These classifiers were Naïve Bayes Tree, C4.5, Bayesian Network, Nearest Neighbor, and K-star algorithms. These selected ML algorithms classify not only flank wear level of the multi-flute end mill but also breakage and chipping. The criterion for selecting the best model depends on classification accuracy and simplicity of the implemented system where the number of features and sensors are kept to minimum in order to increase the efficiency of the online acquisition system. In this paper, all feature selection methods have relatively outperformed the extraction methods using ICA and PCA. In addition, the utilization of simple classifiers such as Nearest Neighbor and K-star classifier has proven to be effective in classifying tools abnormalities in end milling.

KW - Feature selection

KW - Machine Learning

KW - Tool condition monitoring

KW - Tool wear

UR - http://www.scopus.com/inward/record.url?scp=84885993394&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84885993394&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84885993394

SN - 9781627486811

VL - 3

SP - 2378

EP - 2389

BT - 37th International Conference on Computers and Industrial Engineering 2007

ER -