Skip to main navigation Skip to search Skip to main content

Process monitoring of nonlinear industrial process on quality variables based on kernel MPLS

  • Zelin Ren*
  • , Baoran An
  • , Shen Yin
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, a data-driven process monitoring method for quality variables is presented in this paper. Modified partial least squares (MPLS) approach is a kind of lineai multivariate statistical analysis methods, which has a good effeci on process monitoring of linear industrial system. However, it ii not suitable for monitoring nonlinear complex industrial systems Based on it, a nonlinear process monitoring method, kerne modified partial least squares (MKPLS) approach, is obtained by introducing the nuclear method into the MPLS method, in order to detect the fault that can affect quality variables. Firsi of all, the fundamental thoughts of the kernel method are given Then, introduce kernel method to MPLS approach to get MKPLS method. After that, the monitoring effect of MKPLS method on the nonlinear process is tested by nonlinear numerical simulation Finally, MKPLS method is applied to the wastewater treatmen process (WWTP) and process monitoring results are presented in the form of figure.

Original languageEnglish
Title of host publication2018 3rd International Conference on Control and Robotics Engineering, ICCRE 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages280-284
Number of pages5
ISBN (Electronic)9781538666630
DOIs
StatePublished - 8 Jun 2018
Event3rd International Conference on Control and Robotics Engineering, ICCRE 2018 - Nagoya, Japan
Duration: 20 Apr 201823 Apr 2018

Publication series

Name2018 3rd International Conference on Control and Robotics Engineering, ICCRE 2018

Conference

Conference3rd International Conference on Control and Robotics Engineering, ICCRE 2018
Country/TerritoryJapan
CityNagoya
Period20/04/1823/04/18

Keywords

  • Data-driven
  • MKPLS
  • Process monitoring
  • Quality variables

Fingerprint

Dive into the research topics of 'Process monitoring of nonlinear industrial process on quality variables based on kernel MPLS'. Together they form a unique fingerprint.

Cite this