Enhancing chiller energy consumption prediction using gradient boosted trees
DOI:
https://doi.org/10.15282/isse.1.1.2026.13539Keywords:
Chiller energy consumption, Machine learning, Gradient boosting trees, Extreme gradient boosting, Feedforward neural networksAbstract
Chiller systems account for a considerable amount of energy consumption in commercial buildings, which highlights the need of precise energy prediction for operational efficiency and efficient energy management. The standard prediction approaches have difficulties due to the nonlinear interactions between the external variables, the building loads and the chiller operating parameters. The study aims to evaluate the efficiency of Gradient Boosting Trees (GBT) in predicting chiller energy consumption and compare its accuracy with eXtreme Gradient Boosting (XGBoost) and Feedforward Neural Networks (FFNN). The models were constructed based on operational data from a commercial building located in Singapore, with factors such as external temperature, building load and chiller operating parameters. Data pretreatment and feature engineering were performed to improve prediction performance, and grid search was used for hyperparameter optimization. The model evaluation was carried out using the coefficient of determination (R2), root mean square error (RMSE) and mean absolute error (MAE). GBT (R² = 0.9617) had the best prediction accuracy and was slightly better than XGBoost (R² = 0.9586), whilst FFNN exhibited the lowest prediction performance. These results show the effectiveness of GBT in modeling energy consumption of chillers and the potential of tree-based ensemble learning for building energy prediction. Future study can be extended by considering more operational variables and new optimization techniques to further enhance the prediction accuracy.
Downloads
References
[1] G. Hong and N. Seong, “Optimization of the ANN Model for energy consumption prediction of direct-fired absorption chillers for a short-term,” Buildings, vol. 13, no. 10, p. 2526, Oct. 2023, doi: 10.3390/buildings13102526.
[2] T. Sathesh and Y. C. Shih, “Optimized deep learning-based prediction model for chiller performance prediction,” Data Knowl. Eng., vol. 144, p. 102120, Mar. 2023, doi: 10.1016/j.datak.2022.102120.
[3] R. Morcillo-Jimenez, J. Mesa, J. Gómez-Romero, M. A. Vila, and M. J. Martin-Bautista, “Deep learning for prediction of energy consumption: an applied use case in an office building,” Appl. Intell., vol. 54, no. 7, pp. 5813–5825, Apr. 2024, doi: 10.1007/s10489-024-05451-9.
[4] P. Kumar, N. Kamalakshi, and T. Karthick, “Towards sustainable architecture: Machine learning for predicting energy use in buildings,” in Proc. Int. Conf. Adv. Comput. Commun. Appl. Informatics (ACCAI), 2024, pp. 1–6, doi: 10.1109/ACCAI61061.2024.10602461.
[5] I. H. Sarker, “Machine learning: Algorithms, real-world applications and research directions,” SN Comput. Sci., vol. 2, no. 3, p. 160, May 2021, doi: 10.1007/s42979-021-00592-x.
[6] S. Ardabili, L. Abdolalizadeh, C. Mako, B. Torok, and A. Mosavi, “Systematic review of deep learning and machine learning for building energy,” Front. Energy Res., vol. 10, p. 786027, Aug. 2022, doi: 10.3389/fenrg.2022.786027.
[7] M. H. Sulaiman, Z. Mustaffa, and A. S. Abdullah, “Chiller energy prediction in commercial building: A metaheuristic-enhanced deep learning approach,” Energy, vol. 297, p. 131159, Jun. 2024, doi: 10.1016/j.energy.2024.131159.
[8] F. Wahid, R. Ghazali, A. S. Shah, and M. Fayaz, “Prediction of energy consumption in the buildings using multi-layer perceptron and random forest,” Int. J. Adv. Sci. Technol., vol. 101, pp. 13–22, Apr. 2017, doi: 10.14257/ijast.2017.101.02.
[9] S. M. Sojib Ahamed, H. Imran, and I. Saral, “Load prediction of HVAC systems using deep learning,” Glob. Netw., vol. 12, no. 7, pp. 928–972, 2024, doi: 10.5281/zenodo.13134333.
[10] Y. Wang, “Application of deep learning model in building energy consumption prediction,” Comput. Intell. Neurosci., vol. 2022, p. 4835259, Apr. 2022, doi: 10.1155/2022/4835259.
[11] A. Bhansali, N. Narasimhulu, R. Pérez de Prado, P. B. Divakarachari, and D. L. Narayan, “A review on sustainable energy sources using machine learning and deep learning models,” Energies, vol. 16, no. 17, p. 6236, Aug. 2023, doi: 10.3390/en16176236.
[12] Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, May 2015, doi: 10.1038/nature14539.
[13] L. Deng and D. Yu, “Deep learning: Methods and applications,” Found. Trends Signal Process., vol. 7, no. 3–4, pp. 197–387, 2014, doi: 10.1561/2000000039.
[14] J. Schmidhuber, “Deep learning in neural networks: An overview,” Neural Netw., vol. 61, pp. 85–117, Jan. 2015, doi: 10.1016/j.neunet.2014.09.003.
[15] J. Runge and R. Zmeureanu, “Forecasting energy use in buildings using artificial neural networks: A review,” Energies, vol. 12, no. 17, p. 3254, Aug. 2009, doi: 10.3390/en12173254.
[16] U. Kanewala, S. Weerakoon, and R. Nawarathna, “Exploring deep learning and tree-based ensemble models for chiller energy consumption predictions,” in Proc. IEEE 16th Int. Conf. Ind. Inf. Syst. (ICIIS), 2021, pp. 306–311, doi: 10.1109/ICIIS53135.2021.9660662.
[17] W. Li and K. L. E. Law, “Deep learning models for time series forecasting: A review,” IEEE Access, vol. 12, pp. 92306–92327, 2024, doi: 10.1109/ACCESS.2024.3422528.
[18] J. H. Friedman, “Greedy function approximation: A gradient boosting machine,” Ann. Statist., vol. 29, no. 5, pp. 1189–1232, Oct. 2001, doi: 10.1214/aos/1013203451.
[19] Y. Zhang and A. Haghani, “A gradient boosting method to improve travel time prediction,” Transp. Res. Part C Emerg. Technol., vol. 58, pp. 308–324, Sep. 2015, doi: 10.1016/j.trc.2015.02.019.
[20] G. Zheng, Z. Feng, M. Jiang, L. Tan, and Z. Wang, “Predicting the energy consumption of commercial buildings based on deep forest model and its interpretability,” Buildings, vol. 13, no. 9, p. 2162, Aug. 2023, doi: 10.3390/buildings13092162.
[21] T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd ed. New York, NY, USA: Springer, 2009. (Note: Your original text cited a tiny 2-page book review from 2004 instead of the actual world-renowned textbook. I changed this to point directly to the textbook itself).
[22] T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. 22nd ACM SIGKDD Int. Conf. Knowl. Discovery Data Mining, 2016, pp. 785–794, doi: 10.1145/2939672.2939785.
[23] M. Nalluri, M. Pentela, and N. R. Eluri, “A scalable tree boosting system: XG Boost,” Int. J. Res. Stud. Sci. Eng. Technol., vol. 7, no. 12, pp. 35–41, 2020, doi: 10.22259/2349-476X.0712005.
[24] F. Wan, “XGBoost based supply chain fraud detection model,” in Proc. IEEE 2nd Int. Conf. Big Data Artif. Intell. Internet Things Eng. (ICBAIE), 2021, pp. 355–358, doi: 10.1109/ICBAIE52039.2021.9390041.
[25] J. Ye, J. H. Chow, J. Chen, and Z. Zheng, “Stochastic gradient boosted distributed decision trees,” in Proc. 18th ACM Conf. Inf. Knowl. Manage. (CIKM), 2009, pp. 2061–2064, doi: 10.1145/1645953.1646301.
[26] Z. A. Ali, Z. H. Abduljabbar, H. A. Tahir, A. B. Sallow, and S. M. Almufti, “eXtreme gradient boosting algorithm with machine learning: A review,” Acad. J. Nawroz Univ., vol. 12, no. 2, pp. 320–334, May 2023, doi: 10.25007/ajnu.v12n2a1612.
[27] S. Ramraj, N. Uzir, R. Sunil, and S. Banerjee, “Experimenting XGBoost algorithm for prediction and classification of different datasets,” Int. J. Control Theory Appl., vol. 9, no. 40, pp. 651–662, 2016.
[28] K. Hornik, M. Stinchcombe, and H. White, “Multilayer feedforward networks are universal approximators,” Neural Netw., vol. 2, no. 5, pp. 359–366, Jan. 1989, doi: 10.1016/0893-6080(89)90020-8.
[29] S. S. Haykin, Neural Networks and Learning Machines, 3rd ed. Upper Saddle River, NJ, USA: Pearson, 2009.
[30] P. W. Tien, S. Wei, J. Darkwa, C. Wood, and J. K. Calautit, “Machine learning and deep learning methods for enhancing building energy efficiency and indoor environmental quality – A review,” Energy AI, vol. 10, p. 100198, Nov. 2022, doi: 10.1016/j.egyai.2022.100198.
[31] K. Katić, R. Li, J. Verhaart, and W. Zeiler, “Neural network based predictive control of personalized heating systems,” Energy Build., vol. 174, pp. 199–213, Sep. 2018, doi: 10.1016/j.enbuild.2018.06.033.
[32] B. Parizad, H. Ranjbarzadeh, A. Jamali, and H. Khayyam, “An intelligent hybrid machine learning model for sustainable forecasting of home energy demand and electricity price,” Sustainability, vol. 16, no. 6, p. 2328, Mar. 2024, doi: 10.3390/su16062328.
[33] M. Sisode and M. Devare, “A review on machine learning techniques for predictive maintenance in Industry 4.0,” in Proc. Int. Conf. Appl. Mach. Intell. Data Anal. (AMIDA), 2023, pp. 774–833, doi: 10.2991/978-94-6463-136-4_67.
[34] M. Kuhn and K. Johnson, Feature Engineering and Selection: A Practical Approach for Predictive Models. Boca Raton, FL, USA: CRC Press, 2019, doi: 10.1080/00031305.2020.1790217.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 The Author(s)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
