Comparative Analysis of Machine Learning Algorithms for Energy Consumption Forecasting
Article Sidebar
Main Article Content
Abstract: Forecasting energy use has become a crucial component of contemporary smart grid systems, allowing stakeholders to guarantee system dependability, cost effectiveness, and energy efficiency. For the integration of intermittent renewable energy sources, load balancing, and real-time energy management, the capacity to predict power demand is essential. The use and relative effectiveness of five supervised machine learning algorithms Linear Regression, Decision Tree, Random Forest, XGBoost, and Gradient Boosting for predicting short-term building-level energy consumption are examined in this work. In order to train and evaluate models, we carried out a thorough preprocessing and feature engineering procedure using a large dataset that included operational, meteorological, and temporal variables.
Each model was assessed using three key performance metrics: mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R²). Among the models tested, Gradient Boosting achieved the highest accuracy, with an MAE of 575 kWh, RMSE of 851 kWh, and an R² of 0.949, outperforming both traditional and advanced ensemble models.
Our results highlight how well boosting strategies work for energy forecasting jobs and how crucial it is to choose models according to deployment restrictions and data properties. The knowledge gained from this research can help designers create responsive, scalable, and intelligent energy forecasting systems that are appropriate for smart infrastructure.
Downloads
References
Hyndman, R. J. & Athanasopoulos, G. Forecasting: Principles and Practice, 3rd ed., OTexts, 2021.
Breiman, L. “Random Forests,” Mach. Learn. 45, 5 32 (2001).
Friedman, J. H. “Greedy Function Approximation: A Gradient Boosting Machine,” Ann. Stat. 29, 1189 1232 (2001).
Chen, T. & Guestrin, C. “XGBoost: A Scalable Tree Boosting System,” Proc. KDD 2016, 785 794.
Li, K. et al. “Short Term Load Forecasting in Smart Grids: A Combined Deep Learning Approach,” Energy 188, 116–119 (2019).
Lu, M. H. L., Ser, Y. C., Selvachandran, G., Thong, P. H., Cuong, L., Son, L. H., et al. (2022). A comparative study of forecasting electricity consumption using machine learning models. Mathematics, 10(8), 1329.
Bilal, M., Kim, H., Fayaz, M., & Pawar, P. (2022). Comparative analysis of time series forecasting approaches for household electricity consumption prediction. arXiv.
Alawadi, S., Mera, D., Fernández Delgado, M., et al. (2022). A comparison of machine learning algorithms for forecasting indoor temperature in smart buildings. Energy Systems, 13, 689–705.
This work is licensed under a Creative Commons Attribution 4.0 International License.
All articles published in our journal are licensed under CC-BY 4.0, which permits authors to retain copyright of their work. This license allows for unrestricted use, sharing, and reproduction of the articles, provided that proper credit is given to the original authors and the source.