Title: Intelligent building energy consumption prediction system

Authors: Lilei Feng

Addresses: The Second Construction Engineering Co. Ltd., Zhongjian Second Bureau, Zhengzhou City, Henan Province, 450000, China

Abstract: Predicting how well a building consumes energy is an important area of study because it could help make building energy management systems work better. The applied building energy consumption data is used. The proposed method suggests an effective model to predict the energy consumption by a building. The model is good because it fits both the number of hours of heat pain and the amount of energy used during training and tests with R² values higher than 0.97. The gaps in performance between the building energy computer model and field data were, on average, -12.5% less than what was really happening. The intelligence computer model was only 8.65% off when it came to overestimating, which was less than the field data. This paper uses the theoretical framework of artificial neural network for the prediction purpose. The proposed system is limited to our dataset alone at present.

Keywords: artificial neural networks; architectural building design parameters; energy consumption; educational buildings; thermal comfort.

DOI: 10.1504/IJPEC.2026.150635

International Journal of Power and Energy Conversion, 2026 Vol.17 No.1, pp.90 - 102

Received: 30 May 2024
Accepted: 23 Jul 2024
Published online: 18 Dec 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article