An XGBoost-SHAP Model for Energy Demand Prediction With Boruta–Lasso Feature Selection

Energy demand prediction is essential in ensuring national energy security, promoting high-quality economic development, advancing sustainable development, optimizing the energy structure, and achieving dual carbon goals. In recent years, machine learning (ML) algorithms have been extensively used i...

Full description

Saved in:
Bibliographic Details
Main Authors: Yiwen Wang, Weibin Cheng, Yuting Jin, Jifei Li, Yantian Yang, Shaobing Hu
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11098871/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849393794109145088
author Yiwen Wang
Weibin Cheng
Yuting Jin
Jifei Li
Yantian Yang
Shaobing Hu
author_facet Yiwen Wang
Weibin Cheng
Yuting Jin
Jifei Li
Yantian Yang
Shaobing Hu
author_sort Yiwen Wang
collection DOAJ
description Energy demand prediction is essential in ensuring national energy security, promoting high-quality economic development, advancing sustainable development, optimizing the energy structure, and achieving dual carbon goals. In recent years, machine learning (ML) algorithms have been extensively used in energy demand prediction but with poor interpretability. This study proposes an interpretable ML framework for energy demand prediction based on the Boruta-Lasso two-stage feature selection model, extreme gradient boosting (XGBoost) regression model, grid search optimization algorithm, and Shapley additive explanations (SHAP) algorithm. Taking China as an example, the results show that the XGBoost-based model achieves R2, MAE, and RMSE of 0.9975, 0.0108, and 0.0157, respectively, with higher prediction accuracy compared to other models. In addition, the analysis based on the SHAP algorithm shows that gross national income (GNI), gross domestic product (GDP), total population at the year-end, per capita consumption expenditure, and the proportion of added value of the secondary industry are the main drivers of energy demand. The framework established in this study helps to screen influencing factors, predict energy demand, and simultaneously explain the relationship between these factors and energy demand.
format Article
id doaj-art-8a8da6931c3f4bfda5c563b8d87e79c4
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-8a8da6931c3f4bfda5c563b8d87e79c42025-08-20T03:40:17ZengIEEEIEEE Access2169-35362025-01-011313580613582110.1109/ACCESS.2025.359355811098871An XGBoost-SHAP Model for Energy Demand Prediction With Boruta–Lasso Feature SelectionYiwen Wang0https://orcid.org/0009-0002-6214-589XWeibin Cheng1https://orcid.org/0000-0003-4242-1404Yuting Jin2https://orcid.org/0009-0007-7787-8988Jifei Li3https://orcid.org/0009-0008-7986-8721Yantian Yang4https://orcid.org/0009-0002-0924-3503Shaobing Hu5https://orcid.org/0000-0003-2484-7145College of Geophysics and Petroleum Resources, Yangtze University, Wuhan, ChinaCollege of Geophysics and Petroleum Resources, Yangtze University, Wuhan, ChinaCollege of Geophysics and Petroleum Resources, Yangtze University, Wuhan, ChinaCollege of Geophysics and Petroleum Resources, Yangtze University, Wuhan, ChinaCollege of Geophysics and Petroleum Resources, Yangtze University, Wuhan, ChinaCollege of Geophysics and Petroleum Resources, Yangtze University, Wuhan, ChinaEnergy demand prediction is essential in ensuring national energy security, promoting high-quality economic development, advancing sustainable development, optimizing the energy structure, and achieving dual carbon goals. In recent years, machine learning (ML) algorithms have been extensively used in energy demand prediction but with poor interpretability. This study proposes an interpretable ML framework for energy demand prediction based on the Boruta-Lasso two-stage feature selection model, extreme gradient boosting (XGBoost) regression model, grid search optimization algorithm, and Shapley additive explanations (SHAP) algorithm. Taking China as an example, the results show that the XGBoost-based model achieves R2, MAE, and RMSE of 0.9975, 0.0108, and 0.0157, respectively, with higher prediction accuracy compared to other models. In addition, the analysis based on the SHAP algorithm shows that gross national income (GNI), gross domestic product (GDP), total population at the year-end, per capita consumption expenditure, and the proportion of added value of the secondary industry are the main drivers of energy demand. The framework established in this study helps to screen influencing factors, predict energy demand, and simultaneously explain the relationship between these factors and energy demand.https://ieeexplore.ieee.org/document/11098871/Energy demand predictioninterpretabilityfeature selectionBoruta-Lasso modelXGBoostSHAP
spellingShingle Yiwen Wang
Weibin Cheng
Yuting Jin
Jifei Li
Yantian Yang
Shaobing Hu
An XGBoost-SHAP Model for Energy Demand Prediction With Boruta–Lasso Feature Selection
IEEE Access
Energy demand prediction
interpretability
feature selection
Boruta-Lasso model
XGBoost
SHAP
title An XGBoost-SHAP Model for Energy Demand Prediction With Boruta–Lasso Feature Selection
title_full An XGBoost-SHAP Model for Energy Demand Prediction With Boruta–Lasso Feature Selection
title_fullStr An XGBoost-SHAP Model for Energy Demand Prediction With Boruta–Lasso Feature Selection
title_full_unstemmed An XGBoost-SHAP Model for Energy Demand Prediction With Boruta–Lasso Feature Selection
title_short An XGBoost-SHAP Model for Energy Demand Prediction With Boruta–Lasso Feature Selection
title_sort xgboost shap model for energy demand prediction with boruta x2013 lasso feature selection
topic Energy demand prediction
interpretability
feature selection
Boruta-Lasso model
XGBoost
SHAP
url https://ieeexplore.ieee.org/document/11098871/
work_keys_str_mv AT yiwenwang anxgboostshapmodelforenergydemandpredictionwithborutax2013lassofeatureselection
AT weibincheng anxgboostshapmodelforenergydemandpredictionwithborutax2013lassofeatureselection
AT yutingjin anxgboostshapmodelforenergydemandpredictionwithborutax2013lassofeatureselection
AT jifeili anxgboostshapmodelforenergydemandpredictionwithborutax2013lassofeatureselection
AT yantianyang anxgboostshapmodelforenergydemandpredictionwithborutax2013lassofeatureselection
AT shaobinghu anxgboostshapmodelforenergydemandpredictionwithborutax2013lassofeatureselection
AT yiwenwang xgboostshapmodelforenergydemandpredictionwithborutax2013lassofeatureselection
AT weibincheng xgboostshapmodelforenergydemandpredictionwithborutax2013lassofeatureselection
AT yutingjin xgboostshapmodelforenergydemandpredictionwithborutax2013lassofeatureselection
AT jifeili xgboostshapmodelforenergydemandpredictionwithborutax2013lassofeatureselection
AT yantianyang xgboostshapmodelforenergydemandpredictionwithborutax2013lassofeatureselection
AT shaobinghu xgboostshapmodelforenergydemandpredictionwithborutax2013lassofeatureselection