CFD methodology for predicting thermal plume from heat source: Experimental validation and simplified model

Abstract

The prediction of heat transfer in natural convection is crucial for various engineering applications, including building heating and natural ventilation. CFD study of heat transfer in elements with complex geometries, like radiators, increase the computational effort and could turn inviable the study of natural ventilation in large rooms with heating radiators. This study aimed to develop a CFD methodology and validate a simplified model to study natural convection and plumes above heat sources like heating radiators. The model uses porous media to simulate heat sources without compromising thermal plume development in large spaces. It enables cost-effective exploration of solutions, reducing computational costs while accurately modelling thermal plume effects. The CFD model was validated using a full-scale model experimental, ensuring its accuracy and reliability. The experimental measurements showed consistent evolutions for inlet and outlet water temperatures, indicating stable heat transfer processes. The study includes 5 heating scenarios in which the inlet and outlet water temperature (Ti/T0) is varied, namely: 64/58 °C, 67/41 °C, 73/68 °C, 50/35 °C, and 39/29 °C. With the experiments and the CFD results, it was also concluded that the air temperature and velocity profiles in radiators configured in parallel are asymmetrical. The CFD simulations with the simplified model incorporating a porous medium demonstrated the effectiveness of the proposed methodology. Notably, the computational time for the simplified model was reduced by approximately 70 % compared to the detailed model. The developed CFD methodology has potential applications in optimizing natural ventilation systems for different radiators and environmental conditions, contributing to energy efficiency and occupant comfort.