DATA-DRIVEN PREDICTION OF BUILDING ENERGY LOADS
DOI:
https://doi.org/10.69980/pe563v71Keywords:
Building energy prediction, machine learning, gradient boosting, SHAP analysis, load forecastingAbstract
In the modern building world, energy-load prediction is a vital aspect of energy optimisation, demand management and sustainable energy use. This study developed and compared interpretable machine-learning models for predicting hourly building energy consumption across two buildings with different operational characteristics. Multi-year hourly observations were preprocessed through temporal feature engineering, incorporating historical energy consumption, meteorological conditions, and calendar-related variables. The four supervised regression models, namely Linear Regression, Random Forest, Histogram Gradient Boosting, and Extreme Gradient Boosting (XGBoost), were evaluated using an 80:20 chronological train-test split. Predictive performance was assessed using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and the coefficient of determination (R²), while SHapley Additive exPlanations (SHAP) were applied to interpret feature importance. Histogram Gradient Boosting consistently achieved the highest prediction accuracy, obtaining an MAE of 8.063, RMSE of 14.151, MAPE of 5.131%, and R² of 0.9661 for Building A, and an MAE of 3.574, RMSE of 5.481, MAPE of 10.303%, and R² of 0.9752 for Building B. SHAP analysis identified one-hour, daily, and weekly historical energy consumption as the dominant predictors, whereas meteorological variables contributed comparatively less. The results presented indicate that temporal feature engineering and gradient-based ensemble learning can provide an accurate, interpretable and efficient framework for short-term building energy-load forecasting and intelligent energy management.
References
1.Abdunnabi, M., Etiab, N., Nassar, Y. F., El-Khozondar, H. J., & Khargotra, R. (2023). Energy savings strategy for the residential sector in Libya and its impacts on the global environment and the national economy. Advances in Building Energy Research, 17(4), 379-411.
2.Ali, U., Bano, S., Shamsi, M. H., Sood, D., Hoare, C., Zuo, W., ... & O'Donnell, J. (2024). Urban building energy performance prediction and retrofit analysis using a data-driven machine learning approach. Energy and Buildings, 303, 113768.
3.Amasyali, K., & El-Gohary, N. M. (2018). A review of data-driven building energy consumption prediction studies. Renewable and Sustainable Energy Reviews, 81, 1192-1205.
4.Berardi, U. (2017). A cross-country comparison of the building energy consumptions and their trends. Resources, Conservation and Recycling, 123, 230-241.
5.Bourdeau, M., qiang Zhai, X., Nefzaoui, E., Guo, X., & Chatellier, P. (2019). Modeling and forecasting building energy consumption: A review of data-driven techniques. Sustainable Cities and Society, 48, 101533.
6.Chowdhury, B. R. (2025). AI-Powered Forecasting and Optimization of Energy Consumption in the USA: Machine Learning Approaches for Sustainable Urban and Institutional Development. Euro Vantage journals of Artificial intelligence, 2(2), 21-38.
7.González-Torres, M., Pérez-Lombard, L., Coronel, J. F., Maestre, I. R., & Yan, D. (2022). A review on buildings energy information: Trends, end-uses, fuels and drivers. Energy Reports, 8, 626-637.
8.Güneralp, B., Zhou, Y., Ürge-Vorsatz, D., Gupta, M., Yu, S., Patel, P. L., ... & Seto, K. C. (2017). Global scenarios of urban density and its impacts on building energy use through 2050. Proceedings of the National Academy of Sciences, 114(34), 8945-8950.
9.Hong, T., Chen, Y., Luo, X., Luo, N., & Lee, S. H. (2020). Ten questions on urban building energy modeling. Building and Environment, 168, 106508.
10.Le, L. T., Nguyen, H., Dou, J., & Zhou, J. (2019). A comparative study of PSO-ANN, GA-ANN, ICA-ANN, and ABC-ANN in estimating the heating load of buildings’ energy efficiency for smart city planning. Applied Sciences, 9(13), 2630.
11.Levesque, A., Pietzcker, R. C., Baumstark, L., De Stercke, S., Grübler, A., & Luderer, G. (2018). How much energy will buildings consume in 2100? A global perspective within a scenario framework. Energy, 148, 514-527.
12.Mariano, D. (2024). Building energy consumption data (Version 2) [Data set]. Mendeley Data. https://doi.org/10.17632/mzkyh37mtr.2
13.Mishra, A., Lone, H. R., & Mishra, A. (2024). DECODE: Data-driven energy consumption prediction leveraging historical data and environmental factors in buildings. Energy and Buildings, 307, 113950.
14.Molina-Solana, M., Ros, M., Ruiz, M. D., Gómez-Romero, J., & Martín-Bautista, M. J. (2017). Data science for building energy management: A review. Renewable and Sustainable Energy Reviews, 70, 598-609.
15.Natanian, J., Aleksandrowicz, O., & Auer, T. (2019). A parametric approach to optimizing urban form, energy balance and environmental quality: The case of Mediterranean districts. Applied Energy, 254, 113637.
16.Nutkiewicz, A., Yang, Z., & Jain, R. K. (2018). Data-driven Urban Energy Simulation (DUE-S): A framework for integrating engineering simulation and machine learning methods in a multi-scale urban energy modeling workflow. Applied energy, 225, 1176-1189.
17.Pan, Y., & Zhang, L. (2020). Data-driven estimation of building energy consumption with multi-source heterogeneous data. Applied Energy, 268, 114965.
18.Runge, J., & Zmeureanu, R. (2021). A review of deep learning techniques for forecasting energy use in buildings. Energies, 14(3), 608.
19.Seyedzadeh, S., Rahimian, F. P., Glesk, I., & Roper, M. (2018). Machine learning for estimation of building energy consumption and performance: a review. Visualization in Engineering, 6(1), 5.
20.Seyedzadeh, S., Rahimian, F. P., Rastogi, P., & Glesk, I. (2019). Tuning machine learning models for prediction of building energy loads. Sustainable Cities and Society, 47, 101484.
21.Wang, Z., Hong, T., Li, H., & Piette, M. A. (2021). Predicting city-scale daily electricity consumption using data-driven models. Advances in Applied Energy, 2, 100025.
22.Zheng, Z., Zhuang, Z., Lian, Z., & Yu, Y. (2017). Study on building energy load prediction based on monitoring data. Procedia Engineering, 205, 716-723.
23.Zhou, X., Lin, W., Kumar, R., Cui, P., & Ma, Z. (2022). A data-driven strategy using long short term memory models and reinforcement learning to predict building electricity consumption. Applied Energy, 306, 118078.
