A fault diagnosis method in industrial processes with integrated feature space and optimized random forest

Zhenyu Deng, Te Han, Ruonan Liu*, Fengyao Zhi

*Corresponding author for this work

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

1 Citation (Scopus)

Abstract

Many machine learning methods have been successfully applied in fault diagnosis of industrial processes, while integrated feature space has not been fully considered and utilized. In this paper, a fault diagnosis method with integrated feature space and optimized random forest is proposed to realize accurate diagnosis. By means of the integration of static and dynamic information, an embedded & wrapped feature selection process is designed to provide optimal feature combination for classifier. In addition, the optimal feature combination and hyper-parameter optimization are utilized to construct optimized random forest to improve the generalization of model. Experimental results on Tennessee Eastman benchmark show that the proposed method outperforms traditional approaches with accuracy and F1 score that exceed 87% and 88% respectively.

Original languageEnglish
Title of host publication2022 IEEE 31st International Symposium on Industrial Electronics, ISIE 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1170-1173
Number of pages4
ISBN (Electronic)9781665482400
DOIs
Publication statusPublished - 2022
Externally publishedYes
Event31st IEEE International Symposium on Industrial Electronics, ISIE 2022 - Anchorage, United States
Duration: 1 Jun 20223 Jun 2022

Publication series

NameIEEE International Symposium on Industrial Electronics
Volume2022-June

Conference

Conference31st IEEE International Symposium on Industrial Electronics, ISIE 2022
Country/TerritoryUnited States
CityAnchorage
Period1/06/223/06/22

Keywords

  • fault diagnosis
  • industrial process
  • integrated feature space
  • machine learning

Fingerprint

Dive into the research topics of 'A fault diagnosis method in industrial processes with integrated feature space and optimized random forest'. Together they form a unique fingerprint.

Cite this