Assessment of the dynamic thermal behaviour of a test room using computer simulations and experimental measurements

People spend most of the time indoors and adequate comfort level in buildings is fundamental for their well-being and productivity. These issues can be investigated using both simulation models and experimental measurements in dedicated laboratories or in real buildings. The numerical and experimental approaches need to be combined in order to overcome the limitations of each one. This paper presents the application of a detailed numerical model to the CORE-CARE laboratory built at the University of Padua, a test room employed for indoor environmental quality (IEQ) evaluations. The HVAC plant consists of a mechanical ventilation unit and radiant systems for space heating and cooling. Each radiant surface presents different constructive characteristics and can operate independently. The test room is a novelty in this panorama, as it was integrated in an existing building, as a room for regular use. The impossibility of controlling external disturbances makes the dynamic modelling of the room extremely important, for the laboratory control and knowledge. For this reason, the test room behaviour in response to different HVAC system settings and external disturbances is analysed. Two experimental campaigns were carried out both in the heating and cooling seasons, activating one surface at a time, in order to both validate the model and set-up the transient thermal behaviour of the laboratory. In summer, three experimental set-ups were arranged: closed blinds, opened blinds and presence of a thermal dummy. The data analysis highlights that the North surface exhibits the fastest response in both seasons, whereas the floor is the slowest due to its inertia. In summer, the active surface equilibrium temperature is closer to the water supply temperature for responsive and massive surfaces, and it rises with incoming solar radiation and thermal dummy heat gain. The cooled floor enhances the space's thermal stratification. Simulation results show that the model reproduces the system's real dynamics, with a maximum RMSE of 2.3 and 1.9 °C for the North surface and air temperature.