Volume 13 Number 9 (Sep. 2018)
Home > Archive > 2018 > Volume 13 Number 9 (Sep. 2018) >
JSW 2018 Vol.13(9): 506-519 ISSN: 1796-217X
doi: 10.17706/jsw.13.9.506-519

Abnormal Quality Pattern Recognition of Industrial Process Based on Multi-Support Vector Machine

Fengwei Guan1, Lianglun Cheng2*

1 CollegeofC omputer, Guangdong University of Technology, Guang Zhou,510006,Guangdong, China.
2 College of Automation, Guangdong University of Technology, Guangzhou510006, Guangdong, China


Abstract— This paper studies the quality pattern recognitionof industrial processbased on the statistical process control(SPC). An abnormal quality pattern recognition model based on multi-support vector machinewasproposed, which can be used to solve the problem of abnormal pattern recognition in the intelligent manufacturing processfor products.The combination of "one-to-one" and "one-to-many" support vector machine (SVM) classifiersis arranged according to the structure of directed acyclic graphsin the model. At the same time, a structural optimization method was proposed to reduce the cumulative error problem. The model uses the originalfeatures of the datastream of quality. For the support vector machine classifier with low recognition accuracy, the statistical features and shape features form the datastream of quality are integrated withthe original features. Relief algorithm is used to reducethe fusion featuresin order to reduce the consumption caused by increased features. The experimentalresults demonstrate thatthe model improves the accuracy of the recognitionof abnormal patterns, and its structure also has a good time advantage.

Index Terms—Abnormal quality,pattern recognition,SPC,SVM,MSVM.

[PDF]

Cite: Fengwei Guan, Lianglun Cheng, "Abnormal Quality Pattern Recognition of Industrial Process Based on Multi-Support Vector Machine," Journal of Software vol. 13, no. 9, pp. 506-519, 2018.

General Information

ISSN: 1796-217X (Online)
Frequency:  Quarterly
Editor-in-Chief: Prof. Antanas Verikas
Executive Editor: Ms. Yoyo Y. Zhou
Abstracting/ Indexing: DBLP, EBSCO, CNKIGoogle Scholar, ProQuest, INSPEC(IET), ULRICH's Periodicals Directory, WorldCat, etc
E-mail: jsweditorialoffice@gmail.com
  • Apr 26, 2021 News!

    Vol 14, No 4- Vol 14, No 12 has been indexed by IET-(Inspec)     [Click]

  • Nov 18, 2021 News!

    Papers published in JSW Vol 16, No 1- Vol 16, No 6 have been indexed by DBLP   [Click]

  • Dec 24, 2021 News!

     Vol 15, No 1- Vol 15, No 6 has been indexed by IET-(Inspec)   [Click]

  • Jan 04, 2024 News!

    JSW will adopt Article-by-Article Work Flow

  • Dec 06, 2019 News!

    Vol 14, No 1- Vol 14, No 4 has been indexed by EI (Inspec)   [Click]