Issue |
ITM Web Conf.
Volume 70, 2025
2024 2nd International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2024)
|
|
---|---|---|
Article Number | 03011 | |
Number of page(s) | 11 | |
Section | Image Processing and Computer Vision | |
DOI | https://doi.org/10.1051/itmconf/20257003011 | |
Published online | 23 January 2025 |
Research for SARIMA and PatchTSMixer Models on the IEA Monthly Statistics Dataset
Future Tech, South China University of Technology, Guangzhou, Guangdong, China
Corresponding author: 202164690305@mail.scut.edu.cn
The rapid evolution of intelligent algorithms has led to their extensive application in time-series forecasting, particularly in predicting electricity consumption. Accurate forecasting is crucial for energy management, policy-making, and ensuring a stable power supply. However, a significant gap exists in comparing the predictive performance of traditional machine learning methods with advanced deep learning models using real-world datasets. This study aims to address this gap by evaluating and comparing the prediction accuracy of machine learning and deep learning techniques using the Monthly Electricity Statistics dataset from the International Energy Agency (IEA). The research employs a rigorous experimental design, leveraging models such as ARIMA and PatchTSMixer, with an emphasis on model tuning and performance metrics like MAE, MAPE, and RMSE. The findings reveal that deep learning models, particularly PatchTSMixer, outperform traditional machine learning methods in terms of prediction accuracy, demonstrating their superior capability in capturing complex temporal dependencies in electricity consumption data. These results highlight the potential of deep learning models for enhancing predictive accuracy in time-series forecasting, providing valuable insights for future research and practical applications in energy management.
© The Authors, published by EDP Sciences, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.