Cargando…

An Improved Feature Selection Method Based on Random Forest Algorithm for Wind Turbine Condition Monitoring

Feature selection and dimensionality reduction are important for the performance of wind turbine condition monitoring models using supervisory control and data acquisition (SCADA) data. In this paper, an improved random forest algorithm, namely Feature Simplification Random Forest (FS_RF), is propos...

Descripción completa

Detalles Bibliográficos
Autores principales: Li, Guo, Wang, Chensheng, Zhang, Di, Yang, Guang
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8402606/
https://www.ncbi.nlm.nih.gov/pubmed/34451096
http://dx.doi.org/10.3390/s21165654
_version_ 1783745830439616512
author Li, Guo
Wang, Chensheng
Zhang, Di
Yang, Guang
author_facet Li, Guo
Wang, Chensheng
Zhang, Di
Yang, Guang
author_sort Li, Guo
collection PubMed
description Feature selection and dimensionality reduction are important for the performance of wind turbine condition monitoring models using supervisory control and data acquisition (SCADA) data. In this paper, an improved random forest algorithm, namely Feature Simplification Random Forest (FS_RF), is proposed, which is capable of identifying features closely correlated with wind turbine working conditions. The Euclidian distances are employed to distinguish the weight of the same feature among different samples, and its importance is measured by means of the random forest algorithm. The selected features are finally verified by a two-layer gated recurrent unit (GRU) neural network facilitating condition monitoring. The experimental results demonstrate the capacity and effectiveness of the proposed method for wind turbine condition monitoring.
format Online
Article
Text
id pubmed-8402606
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-84026062021-08-29 An Improved Feature Selection Method Based on Random Forest Algorithm for Wind Turbine Condition Monitoring Li, Guo Wang, Chensheng Zhang, Di Yang, Guang Sensors (Basel) Article Feature selection and dimensionality reduction are important for the performance of wind turbine condition monitoring models using supervisory control and data acquisition (SCADA) data. In this paper, an improved random forest algorithm, namely Feature Simplification Random Forest (FS_RF), is proposed, which is capable of identifying features closely correlated with wind turbine working conditions. The Euclidian distances are employed to distinguish the weight of the same feature among different samples, and its importance is measured by means of the random forest algorithm. The selected features are finally verified by a two-layer gated recurrent unit (GRU) neural network facilitating condition monitoring. The experimental results demonstrate the capacity and effectiveness of the proposed method for wind turbine condition monitoring. MDPI 2021-08-22 /pmc/articles/PMC8402606/ /pubmed/34451096 http://dx.doi.org/10.3390/s21165654 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Li, Guo
Wang, Chensheng
Zhang, Di
Yang, Guang
An Improved Feature Selection Method Based on Random Forest Algorithm for Wind Turbine Condition Monitoring
title An Improved Feature Selection Method Based on Random Forest Algorithm for Wind Turbine Condition Monitoring
title_full An Improved Feature Selection Method Based on Random Forest Algorithm for Wind Turbine Condition Monitoring
title_fullStr An Improved Feature Selection Method Based on Random Forest Algorithm for Wind Turbine Condition Monitoring
title_full_unstemmed An Improved Feature Selection Method Based on Random Forest Algorithm for Wind Turbine Condition Monitoring
title_short An Improved Feature Selection Method Based on Random Forest Algorithm for Wind Turbine Condition Monitoring
title_sort improved feature selection method based on random forest algorithm for wind turbine condition monitoring
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8402606/
https://www.ncbi.nlm.nih.gov/pubmed/34451096
http://dx.doi.org/10.3390/s21165654
work_keys_str_mv AT liguo animprovedfeatureselectionmethodbasedonrandomforestalgorithmforwindturbineconditionmonitoring
AT wangchensheng animprovedfeatureselectionmethodbasedonrandomforestalgorithmforwindturbineconditionmonitoring
AT zhangdi animprovedfeatureselectionmethodbasedonrandomforestalgorithmforwindturbineconditionmonitoring
AT yangguang animprovedfeatureselectionmethodbasedonrandomforestalgorithmforwindturbineconditionmonitoring
AT liguo improvedfeatureselectionmethodbasedonrandomforestalgorithmforwindturbineconditionmonitoring
AT wangchensheng improvedfeatureselectionmethodbasedonrandomforestalgorithmforwindturbineconditionmonitoring
AT zhangdi improvedfeatureselectionmethodbasedonrandomforestalgorithmforwindturbineconditionmonitoring
AT yangguang improvedfeatureselectionmethodbasedonrandomforestalgorithmforwindturbineconditionmonitoring