Abstract
A deviation between the calculated and the measured data of energy use in buildings (known as performance gap) has been extensively reported. The aim of this study is to identify the underlying causes of the deviation between the measurements and simulation data of the temperature profiles in a test cell facility located in Trondheim, Norway. Simulated and measured temperatures were compared and the occupants’ feedback was analysed in detail. The results show that the deviation between the simulations and the measurements was smaller for the air and operative temperature than that for the surface and mean radiative temperature. The underlying reasons for the deviations can mainly be attributed to the missing ability to consider temperature stratification and the use of constant air velocity in the simulation program, and errors in splitting global horizontal radiation in its direct and diffuse components.