Mon, Jun 30, 2025
[Archive]
Abstract: (86 Views)
Traditional energy performance analysis methods have become obsolete because of the development of novel swarm-based optimization methods. The research examines the potential of two hybrid approaches, notably genetic algorithm (GA) and grey wolf optimization (GWO), for enhancing the neural evaluation of cooling load (CL) in green buildings. To accomplish such an objective, preparing the necessary dataset considers eight CL-influencing elements, such as relative compactness, surface area, glazing area distribution, wall area, overall height, roof area, and orientation. A population-based analysis is conducted using the best-fitting architectures of every approach. According to the findings, using both GA and GWO algorithms increased neural network accuracy. The analysis outcomes indicate that the GA model was the most incredible ANFIS model. Significantly, the GA-ANFIS model had a greater 
amount of 0.98 and the lowest RMSE amount of 0.09 among the two models examined. The GWO-ANFIS approach achieved
amount of 0.98 and the lowest RMSE amount of 0.09 among the two models examined. The GWO-ANFIS approach achieved amounts of 0.95 and RMSE amounts of 0.15, indicating that GA-ANFIS offers superior performance.
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
