Random forest-based nonlinear improved feature extraction and selection for fault classification

Radhia Fezai, Kais Bouzrara, Majdi Mansouri, Hazem Nounou, Mohamed Nounou, Mohamed Trabelsi

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

1 Citation (Scopus)

Abstract

In this paper, Interval Gaussian Process Regression (IGPR)-based Random Forest (RF) proposed for fault detection and diagnosis (FDD) due to its effectiveness in handling uncertain industrial process data, which are often with high nonlinearities and strong correlations. This technique aims to extract the features from raw data using IGPR technique. Then, the interval mean vector and the interval variance matrix obtained from IGPR technique are used as inputs to the Random Forest (RF) classifier. The results show the effectiveness of the features and the classifiers in detection of faults of Wind Energy Conversion (WEC) Systems.

Original languageEnglish
Title of host publication18th IEEE International Multi-Conference on Systems, Signals and Devices, SSD 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages601-606
Number of pages6
ISBN (Electronic)9781665414937
DOIs
Publication statusPublished - 22 Mar 2021
Externally publishedYes
Event18th IEEE International Multi-Conference on Systems, Signals and Devices, SSD 2021 - Monastir, Tunisia
Duration: 22 Mar 202125 Mar 2021

Publication series

Name18th IEEE International Multi-Conference on Systems, Signals and Devices, SSD 2021

Conference

Conference18th IEEE International Multi-Conference on Systems, Signals and Devices, SSD 2021
Country/TerritoryTunisia
CityMonastir
Period22/03/2125/03/21

Keywords

  • Fault Detection and Diagnosis (FDD)
  • Feature Extraction and Selection
  • Interval Gaussian Process Regression (IGPR)
  • Interval-Valued Data
  • Random Forest (RF)
  • Wind Energy Conversion (WEC) Systems

Fingerprint

Dive into the research topics of 'Random forest-based nonlinear improved feature extraction and selection for fault classification'. Together they form a unique fingerprint.

Cite this