Operator-based nonlinear fault detection and fault tolerance control for microreactor with generalized Gaussian function

Yoshiki Ogihara, Mingcong Deng, Yuichi Noge

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

2 Citations (Scopus)

Abstract

In this paper, operator-based method of fault detection and fault tolerance control using generalized Gaussian function as the kernel function of Support Vector Machine (SVM) is proposed for microreactor units actuated by Peltier device. In details, after describing a mathematical model of the microreactor, an operator-based robust fault tolerant control system is designed for this process with output sensor fault. Then real-Time fault detection scheme is presented by using SVM, where generalized Gaussian function is adopted as the kernel function. The effectiveness of the proposed method is confirmed by simulation and experimental result.

Original languageEnglish
Title of host publication1st International Conference on Industrial Artificial Intelligence, IAI 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728135939
DOIs
Publication statusPublished - Jul 2019
Externally publishedYes
Event1st International Conference on Industrial Artificial Intelligence, IAI 2019 - Shenyang, China
Duration: Jul 22 2019Jul 26 2019

Publication series

Name1st International Conference on Industrial Artificial Intelligence, IAI 2019

Conference

Conference1st International Conference on Industrial Artificial Intelligence, IAI 2019
Country/TerritoryChina
CityShenyang
Period7/22/197/26/19

All Science Journal Classification (ASJC) codes

  • Process Chemistry and Technology
  • Artificial Intelligence
  • Computer Science Applications
  • Industrial and Manufacturing Engineering
  • Control and Optimization

Fingerprint

Dive into the research topics of 'Operator-based nonlinear fault detection and fault tolerance control for microreactor with generalized Gaussian function'. Together they form a unique fingerprint.

Cite this