Fault detection and diagnosis with random forest feature extraction and variable importance methods

The ever-present drive to safer, more cost-effective and cleaner processes motivates the exploration of a variety of process monitoring methods. In the domain of data-driven approaches, random forest models present a nonlinear framework. Random forest models consist of ensembles of classification an...

Full description

Bibliographic Details
Main Authors: Aldrich, Chris, Auret, L.
Other Authors: C Aldrich
Format: Conference Paper
Published: Elsevier 2010
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/27554
_version_ 1848752295784742912
author Aldrich, Chris
Auret, L.
author2 C Aldrich
author_facet C Aldrich
Aldrich, Chris
Auret, L.
author_sort Aldrich, Chris
building Curtin Institutional Repository
collection Online Access
description The ever-present drive to safer, more cost-effective and cleaner processes motivates the exploration of a variety of process monitoring methods. In the domain of data-driven approaches, random forest models present a nonlinear framework. Random forest models consist of ensembles of classification and regression trees in which the model response is determined by voting committees of independent binary decision trees. Data-driven approaches to fault diagnosis often involve summarizing potentially large numbers of process variables in lower dimensional diagnostic sequences. Random forest feature extraction allows for the monitoring of process in feature and residual spaces, while random forest variable importance measures can potentially be used to identify process variables contribution to fault conditions. In this study, a framework for diagnosing steady state faults with random forests is proposed and demonstrated with a simple nonlinear system and the benchmark Tennessee Eastman process.
first_indexed 2025-11-14T08:06:21Z
format Conference Paper
id curtin-20.500.11937-27554
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T08:06:21Z
publishDate 2010
publisher Elsevier
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-275542023-01-13T07:56:31Z Fault detection and diagnosis with random forest feature extraction and variable importance methods Aldrich, Chris Auret, L. C Aldrich L Auret fault diagnosis feature extraction multivariate - statistical process control Random forest models variable importance The ever-present drive to safer, more cost-effective and cleaner processes motivates the exploration of a variety of process monitoring methods. In the domain of data-driven approaches, random forest models present a nonlinear framework. Random forest models consist of ensembles of classification and regression trees in which the model response is determined by voting committees of independent binary decision trees. Data-driven approaches to fault diagnosis often involve summarizing potentially large numbers of process variables in lower dimensional diagnostic sequences. Random forest feature extraction allows for the monitoring of process in feature and residual spaces, while random forest variable importance measures can potentially be used to identify process variables contribution to fault conditions. In this study, a framework for diagnosing steady state faults with random forests is proposed and demonstrated with a simple nonlinear system and the benchmark Tennessee Eastman process. 2010 Conference Paper http://hdl.handle.net/20.500.11937/27554 Elsevier restricted
spellingShingle fault diagnosis
feature extraction
multivariate - statistical process control
Random forest models
variable importance
Aldrich, Chris
Auret, L.
Fault detection and diagnosis with random forest feature extraction and variable importance methods
title Fault detection and diagnosis with random forest feature extraction and variable importance methods
title_full Fault detection and diagnosis with random forest feature extraction and variable importance methods
title_fullStr Fault detection and diagnosis with random forest feature extraction and variable importance methods
title_full_unstemmed Fault detection and diagnosis with random forest feature extraction and variable importance methods
title_short Fault detection and diagnosis with random forest feature extraction and variable importance methods
title_sort fault detection and diagnosis with random forest feature extraction and variable importance methods
topic fault diagnosis
feature extraction
multivariate - statistical process control
Random forest models
variable importance
url http://hdl.handle.net/20.500.11937/27554