Predicting Long-Term Electricity Consumption Using Time Series Data: Use Case of the UK Electricity Data

De Silva, Samudra Nimesh, Mishra, Bhupesh Kumar, Sayers, William ORCID: https://orcid.org/0000-0003-1677-4409 and Loukil, Zainab ORCID: https://orcid.org/0000-0003-2731-7051 (2025) Predicting Long-Term Electricity Consumption Using Time Series Data: Use Case of the UK Electricity Data. In: Intelligent Systems with Applications in Communications, Computing and IoT (ICISCCI 2024). Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering Ser., 621 (621). Springer, Switzerland, pp. 37-58. ISBN 9783031926136

Text 15214 De Silva, Samudra Nimesh et al. (2025) Predicting Long Term Electricity Consumption.pdf - Accepted Version Restricted to Repository staff only Available under License All Rights Reserved. Download (4MB)

Abstract

In the modern era, the United Kingdom (UK) is moving towards a future enriched with sustainability and energy efficiency. Thus, accurate electricity consumption predictions are one of the major factors in the UK. This research focuses on electricity consumption in the UK for the next year using historical data and analyzing them through time series analysis. Traditional statistical methods such as Autoregressive Integrated Moving Average (ARIMA), and Seasonal Autoregressive Moving Average (SARIMA) and machine learning algorithms such as XGBoost, Linear Trees and Prophet are used in this research to predict the consumption patterns of electricity in the UK. Moreover, this study will further focus on evaluating the performances and the behavioral patterns of each model in forecasting electricity consumption by developing a comparative analysis that highlights the strengths and weaknesses of each model. The main objective of the following research is to create a model which can forecast the consumption of electricity for the next year in the UK whilst identifying the consumption patterns of electricity in the UK. Hence, this analysis encompasses valuable information for energy providers, consumers and policymakers to enhance energy efficiency and maintain energy sustainability in the UK. Furthermore, by this analysis it was determined that Prophet model is the best model to be used in predicting the electricity consumption for next year in the UK as it has generated significant results compared to other models making the Mean Absolute Percentage Error (MAPE) 0.14% and the Root Mean Squared Error (RMSE) 4409.71.

Item Type:	Book Section
Uncontrolled Keywords:	Machine Learning; Time Series; Seasonal Patterns; Electricity; ARIMA; SARIMA; Prophet; XGBoost
Subjects:	H Social Sciences > HA Statistics Q Science > QA Mathematics > QA76 Computer software > QA76.9 Other topics > QA76.9.B45 Big data
Divisions:	Schools and Research Institutes > School of Business, Computing and Social Sciences
Depositing User:	Kamila Niekoraniec
Date Deposited:	19 Aug 2025 14:49
Last Modified:	20 Aug 2025 11:15
URI:	https://eprints.glos.ac.uk/id/eprint/15214

University Staff: Request a correction | Repository Editors: Update this record