Mon, Jun 30, 2025
[Archive]
School of Electrical and Electronic Engineering, Universiti Sains Malaysia (USM), Nibong Tebal 14300, Penang, Malaysia
Abstract: (76 Views)
Planning, management, and energy conservation all benefit from accurate predictions of building energy usage. The secret to ensuring energy systems' performance and sustainability is continuously improving and enhancing forecasting models' effectiveness. In this context, the current research presents, after studying and evaluating several kinds of HL forecasting models, a new enhanced hybrid approach of machine learning application for predicting residential buildings' heating load (HL). The suggested hybrid model, GWO-ANIFS, combines the support vector regression (GWO) and group technique of ANFIS models. The forecasting models used the building’s technical characteristics as input factors, and the HL was chosen as the network's output variable. The findings showed that the suggested ANFIS approach with a 100-person population size was the best approach for forecasting building energy because it had the highest R2 (0.97905 and 0.9789) and the lowest error amunts in the forms of MSE (0.012433), RMSE (0.1115), and MAE (0.088128) for predicting HL.
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
