TY - GEN
T1 - A Novel Statistical Method for Extrapolating External Causality to Electricity Demand Growth
AU - Xu, Xinyi
AU - Bao, Yucheng
AU - Fang, Lurui
AU - Chen, Xiaoyang
AU - Lim, Eng Gee
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.
PY - 2025
Y1 - 2025
N2 - Electricity consumption is a vital index significantly impacting the efficient operation of power transmission systems. This research explores the correlation between external causality, such as urban population, and electricity demand increase by processing relevant data and constructing models, offering insights for long-term electricity demand estimation and related network planning formulation at the national level. By conducting feature extraction and analysis, this study reveals the inner correlation between these variables using Pearson and Spearman correlation coefficients, regardless of the disparities among different countries. Utilizing statistical modeling, this research analyzes data on electricity demand growth across 25 developed regions. After integrating the data, an applicable regression model – XGBoost Regressor is developed to predict national electricity consumption, incorporating evaluation metrics for comparison and accuracy validation. Compared with other models, the forecasting of the XGBoost Regressor is more stable, achieving satisfactory accuracy on the given samples without overly biased predictions.
AB - Electricity consumption is a vital index significantly impacting the efficient operation of power transmission systems. This research explores the correlation between external causality, such as urban population, and electricity demand increase by processing relevant data and constructing models, offering insights for long-term electricity demand estimation and related network planning formulation at the national level. By conducting feature extraction and analysis, this study reveals the inner correlation between these variables using Pearson and Spearman correlation coefficients, regardless of the disparities among different countries. Utilizing statistical modeling, this research analyzes data on electricity demand growth across 25 developed regions. After integrating the data, an applicable regression model – XGBoost Regressor is developed to predict national electricity consumption, incorporating evaluation metrics for comparison and accuracy validation. Compared with other models, the forecasting of the XGBoost Regressor is more stable, achieving satisfactory accuracy on the given samples without overly biased predictions.
KW - Correlation Analysis
KW - Electricity Consumption
KW - Electricity Demand Prediction
KW - Regression Model
UR - http://www.scopus.com/inward/record.url?scp=105000831432&partnerID=8YFLogxK
U2 - 10.1007/978-981-96-2456-0_15
DO - 10.1007/978-981-96-2456-0_15
M3 - Conference Proceeding
AN - SCOPUS:105000831432
SN - 9789819624553
T3 - Lecture Notes in Electrical Engineering
SP - 131
EP - 143
BT - Proceedings of 2024 International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024 - Volume 1
A2 - Wen, Fushuan
A2 - Liu, Haoming
A2 - Wen, Huiqing
A2 - Wang, Shunli
PB - Springer Science and Business Media Deutschland GmbH
T2 - 2nd International Conference on Smart Electrical Grid and Renewable Energy, SEGRE 2024
Y2 - 9 August 2024 through 12 August 2024
ER -
