A transient model for radiant heating and cooling terminal heat exchangers applied to radiant floors and ceiling panels

Abstract

Renewable energy technologies for heating and cooling can be often optimized using low temperatures in the terminal heat exchangers for heating and the opposite for cooling. For example, thermal and electrical driven heat pumps have higher coefficient of performance (COP) for low impulse heating temperature and relatively high cooling temperature. If solar energy is directly used for heating purposes also a low temperature is necessary. Using this range of temperatures, a large exchange area is needed in order to obtain the desired conditions. In this sense, radiant floors and ceilings can be used with the objective to reduce the impulse water temperature in winter and increase it in summer to obtain high energy savings. However, a careful design and optimal control strategy are important to reach expected energy savings. Therefore, a model capable to capture transient effects, system control strategies, and it’s coupling with building energy simulation, is of importance. A transient numerical model for radiant floors and ceilings is presented and validated. The model has been implemented in RDmes online web platform and can solve both steady state and transient situations for sizing and predicting respectively. The radiant floor model has been developed to be used in the framework of the IEA-Task44 and the radiant ceiling has been employed in the FREDSOL project. The model developed is based on the composite fin model for radiant floor described by Kilkis et. al. (1994) coupled with a multi layer model and a step-by-step algorithm. The multi layer model solves the transient one-dimensional conduction behavior of the different layers. The two models are coupled considering the heat flow from the pipe as a sink source term in a typical transient conduction problem. The similar concept was used in the collector model presented by Cadafalch (2009). The step-by-step algorithm solves the fluid flow in one dimension. In the paper, an explanation of the mathematical formulation and numerical algorithm is provided in detail. A validation has been realized by means of experimental data comparisons from other references. Computational results have been analyzed and compared with other models presented in the test cases summarized in.