Employment of a Radial Basis Function Model for Predicting the Heating Load of Construction

Abstract

Nowadays, the main focus of current research and practice is to prioritize energy-efficient building management. Therefore, this pressing need pushes the current study to offer an allaround solution by fusing advanced enhancement frameworks with the meticulous forecasting of heating load (HL). This research activity takes place against the challenging backdrop of complex heating, ventilation, and air conditioning (HVAC) systems where energy optimization encompasses many variables requiring thorough investigation with creative methodologies. This work highlights that HL prediction could play a great role in improvements to enhance HVAC system performances, energy efficiencies, and thereby cost benefits. The innovative approaches presented in this research consist of integrating 2 advanced optimizers, namely an Improved Manta-Ray Foraging Optimizer (IMRFO) and a Population-based Vortex Search Algorithm (PVSA), with a Radial Basis Function (RBF). The overall objective is to boost the precision of HL predictions and simplify the optimization process of HVAC systems. The fact that this research is directed toward the goal of finding energy efficiency and cost-effectiveness, and generally toward the objective of improving the sustainability of the environment in building operation, speaks to the very central role that accurate HL prediction will play. These validations also prove that the RBPV model is the most outstanding regarding real-world applicability and accuracy. It attains an outstanding maximum R^2 train value of 0.992 , indicating a high degree of explanatory power and an impressively low RMSE_train value of 0.896, signifying minimal prediction errors in comparison to other frameworks.

Keywords

• heating load
• heating energy consumption
• radial basis function
• populationbased vortex search algorithm
• improved manta-ray foraging optimizer