No Cover Image

Journal article 195 views 65 downloads

Innovative framework for accurate and transparent forecasting of energy consumption: A fusion of feature selection and interpretable machine learning

Hamid Eskandari Orcid Logo, Hassan Saadatmand, Muhammad Ramzan, Mobina Mousapour

Applied Energy, Volume: 366, Start page: 123314

Swansea University Author: Hamid Eskandari Orcid Logo

  • 66476.VoR.pdf

    PDF | Version of Record

    © 2024 The Authors. This is an open access article under the CC BY license.

    Download (9.84MB)

Check full text

DOI (Published version): 10.1016/j.apenergy.2024.123314

Abstract

The study presents a novel framework integrating feature selection (FS) and machine learning (ML) techniques to forecast inland national energy consumption (EC) in the United Kingdom across all energy sources. This innovative framework strategically combines three FS approaches with five interpretab...

Full description

Published in: Applied Energy
ISSN: 0306-2619
Published: Elsevier BV 2024
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa66476
first_indexed 2024-05-20T09:50:45Z
last_indexed 2024-11-25T14:18:14Z
id cronfa66476
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2024-05-20T10:52:49.1376603</datestamp><bib-version>v2</bib-version><id>66476</id><entry>2024-05-20</entry><title>Innovative framework for accurate and transparent forecasting of energy consumption: A fusion of feature selection and interpretable machine learning</title><swanseaauthors><author><sid>d2a47b056b55373889a9d19d2924f634</sid><ORCID>0000-0002-5515-9399</ORCID><firstname>Hamid</firstname><surname>Eskandari</surname><name>Hamid Eskandari</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2024-05-20</date><deptcode>CBAE</deptcode><abstract>The study presents a novel framework integrating feature selection (FS) and machine learning (ML) techniques to forecast inland national energy consumption (EC) in the United Kingdom across all energy sources. This innovative framework strategically combines three FS approaches with five interpretable ML models using Shapley Additive Explanations (SHAP), with the dual goal of enhancing accuracy and transparency in EC predictions. By meticulously selecting the most pertinent features from diverse features&#x2014;including meteorological conditions, socioeconomic parameters, and historical consumption patterns of different primary fuels&#x2014;the proposed framework enhances the robustness of the forecasting model. This is achieved through benchmarking three FS approaches: ensemble filter, wrapper, and a hybrid ensemble filter-wrapper. In addition, we introduce a novel ensemble filter FS, synthesizing outcomes from multiple base FS methods to make well-informed decisions about feature retention. Experimental results underscore the efficacy of integrating both wrapper and ensemble filter-wrapper FS approaches with interpretable ML models, ensuring the forecasting process remains comprehensible and interpretable while utilizing a manageable number of features (four to eight). In addition, experimental results indicate that different feature subsets are usually selected for each combined FS approach and ML model. This study not only demonstrates the framework's capability to provide accurate forecasts but also establishes it as a valuable tool for policymakers and energy analysts.</abstract><type>Journal Article</type><journal>Applied Energy</journal><volume>366</volume><journalNumber/><paginationStart>123314</paginationStart><paginationEnd/><publisher>Elsevier BV</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>0306-2619</issnPrint><issnElectronic/><keywords>Energy consumption forecasting; Interpretable machine learning; Ensemble feature selection; Wrapper feature selection; Shapley analysis</keywords><publishedDay>15</publishedDay><publishedMonth>7</publishedMonth><publishedYear>2024</publishedYear><publishedDate>2024-07-15</publishedDate><doi>10.1016/j.apenergy.2024.123314</doi><url/><notes/><college>COLLEGE NANME</college><department>Management School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>CBAE</DepartmentCode><institution>Swansea University</institution><apcterm>SU Library paid the OA fee (TA Institutional Deal)</apcterm><funders/><projectreference/><lastEdited>2024-05-20T10:52:49.1376603</lastEdited><Created>2024-05-20T10:48:41.4588852</Created><path><level id="1">Faculty of Humanities and Social Sciences</level><level id="2">School of Management - Business Management</level></path><authors><author><firstname>Hamid</firstname><surname>Eskandari</surname><orcid>0000-0002-5515-9399</orcid><order>1</order></author><author><firstname>Hassan</firstname><surname>Saadatmand</surname><order>2</order></author><author><firstname>Muhammad</firstname><surname>Ramzan</surname><order>3</order></author><author><firstname>Mobina</firstname><surname>Mousapour</surname><order>4</order></author></authors><documents><document><filename>66476__30398__2db22f7669a245afb2817f5b5fb7d9dd.pdf</filename><originalFilename>66476.VoR.pdf</originalFilename><uploaded>2024-05-20T10:51:26.2262471</uploaded><type>Output</type><contentLength>10316140</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>&#xA9; 2024 The Authors. This is an open access article under the CC BY license.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>http://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2024-05-20T10:52:49.1376603 v2 66476 2024-05-20 Innovative framework for accurate and transparent forecasting of energy consumption: A fusion of feature selection and interpretable machine learning d2a47b056b55373889a9d19d2924f634 0000-0002-5515-9399 Hamid Eskandari Hamid Eskandari true false 2024-05-20 CBAE The study presents a novel framework integrating feature selection (FS) and machine learning (ML) techniques to forecast inland national energy consumption (EC) in the United Kingdom across all energy sources. This innovative framework strategically combines three FS approaches with five interpretable ML models using Shapley Additive Explanations (SHAP), with the dual goal of enhancing accuracy and transparency in EC predictions. By meticulously selecting the most pertinent features from diverse features—including meteorological conditions, socioeconomic parameters, and historical consumption patterns of different primary fuels—the proposed framework enhances the robustness of the forecasting model. This is achieved through benchmarking three FS approaches: ensemble filter, wrapper, and a hybrid ensemble filter-wrapper. In addition, we introduce a novel ensemble filter FS, synthesizing outcomes from multiple base FS methods to make well-informed decisions about feature retention. Experimental results underscore the efficacy of integrating both wrapper and ensemble filter-wrapper FS approaches with interpretable ML models, ensuring the forecasting process remains comprehensible and interpretable while utilizing a manageable number of features (four to eight). In addition, experimental results indicate that different feature subsets are usually selected for each combined FS approach and ML model. This study not only demonstrates the framework's capability to provide accurate forecasts but also establishes it as a valuable tool for policymakers and energy analysts. Journal Article Applied Energy 366 123314 Elsevier BV 0306-2619 Energy consumption forecasting; Interpretable machine learning; Ensemble feature selection; Wrapper feature selection; Shapley analysis 15 7 2024 2024-07-15 10.1016/j.apenergy.2024.123314 COLLEGE NANME Management School COLLEGE CODE CBAE Swansea University SU Library paid the OA fee (TA Institutional Deal) 2024-05-20T10:52:49.1376603 2024-05-20T10:48:41.4588852 Faculty of Humanities and Social Sciences School of Management - Business Management Hamid Eskandari 0000-0002-5515-9399 1 Hassan Saadatmand 2 Muhammad Ramzan 3 Mobina Mousapour 4 66476__30398__2db22f7669a245afb2817f5b5fb7d9dd.pdf 66476.VoR.pdf 2024-05-20T10:51:26.2262471 Output 10316140 application/pdf Version of Record true © 2024 The Authors. This is an open access article under the CC BY license. true eng http://creativecommons.org/licenses/by/4.0/
title Innovative framework for accurate and transparent forecasting of energy consumption: A fusion of feature selection and interpretable machine learning
spellingShingle Innovative framework for accurate and transparent forecasting of energy consumption: A fusion of feature selection and interpretable machine learning
Hamid Eskandari
title_short Innovative framework for accurate and transparent forecasting of energy consumption: A fusion of feature selection and interpretable machine learning
title_full Innovative framework for accurate and transparent forecasting of energy consumption: A fusion of feature selection and interpretable machine learning
title_fullStr Innovative framework for accurate and transparent forecasting of energy consumption: A fusion of feature selection and interpretable machine learning
title_full_unstemmed Innovative framework for accurate and transparent forecasting of energy consumption: A fusion of feature selection and interpretable machine learning
title_sort Innovative framework for accurate and transparent forecasting of energy consumption: A fusion of feature selection and interpretable machine learning
author_id_str_mv d2a47b056b55373889a9d19d2924f634
author_id_fullname_str_mv d2a47b056b55373889a9d19d2924f634_***_Hamid Eskandari
author Hamid Eskandari
author2 Hamid Eskandari
Hassan Saadatmand
Muhammad Ramzan
Mobina Mousapour
format Journal article
container_title Applied Energy
container_volume 366
container_start_page 123314
publishDate 2024
institution Swansea University
issn 0306-2619
doi_str_mv 10.1016/j.apenergy.2024.123314
publisher Elsevier BV
college_str Faculty of Humanities and Social Sciences
hierarchytype
hierarchy_top_id facultyofhumanitiesandsocialsciences
hierarchy_top_title Faculty of Humanities and Social Sciences
hierarchy_parent_id facultyofhumanitiesandsocialsciences
hierarchy_parent_title Faculty of Humanities and Social Sciences
department_str School of Management - Business Management{{{_:::_}}}Faculty of Humanities and Social Sciences{{{_:::_}}}School of Management - Business Management
document_store_str 1
active_str 0
description The study presents a novel framework integrating feature selection (FS) and machine learning (ML) techniques to forecast inland national energy consumption (EC) in the United Kingdom across all energy sources. This innovative framework strategically combines three FS approaches with five interpretable ML models using Shapley Additive Explanations (SHAP), with the dual goal of enhancing accuracy and transparency in EC predictions. By meticulously selecting the most pertinent features from diverse features—including meteorological conditions, socioeconomic parameters, and historical consumption patterns of different primary fuels—the proposed framework enhances the robustness of the forecasting model. This is achieved through benchmarking three FS approaches: ensemble filter, wrapper, and a hybrid ensemble filter-wrapper. In addition, we introduce a novel ensemble filter FS, synthesizing outcomes from multiple base FS methods to make well-informed decisions about feature retention. Experimental results underscore the efficacy of integrating both wrapper and ensemble filter-wrapper FS approaches with interpretable ML models, ensuring the forecasting process remains comprehensible and interpretable while utilizing a manageable number of features (four to eight). In addition, experimental results indicate that different feature subsets are usually selected for each combined FS approach and ML model. This study not only demonstrates the framework's capability to provide accurate forecasts but also establishes it as a valuable tool for policymakers and energy analysts.
published_date 2024-07-15T08:30:46Z
_version_ 1821393548189106176
score 11.04748

Similar Items