Nowadays it is advisable for buildings to be thermally resilient, so as to passively resist to the daily external air temperature fluctuations, reducing the interventions of air conditioning systems to a minimum. The decrement factor and time lag are parameters notoriously used to evaluate the responses of the building envelope to periodic oscillations in external air temperature. However, in their traditional definition, these parameters do not consider the effects of internal elements, such as partition walls and slabs, which instead influence the passive thermal behaviour of the building. For this reason, a choice of the external wall stratigraphy, made exclusively on the basis of the traditional decrement factor and time lag, seems to be simplistic. In this paper, problems concerning the passive behaviour of buildings, in case of sinusoidal outdoor air temperature variations, are investigated. Within a lumped-capacitance scheme and using the heat transfer matrix method, the problem of the determination of a wall stratigraphy minimizing the decrement factor for given values of the overall thermal resistance and capacity, usable for the realization of both external and internal walls of a given room, is solved. After developing an analysis on the influence of the internal walls and slabs on the room thermal balance, the resistance-capacity distribution, within the external wall, minimizing the decrement factor is determined, for assigned thermal resistances and capacities of internal elements and the external wall. Results indicate that there is no resistance-capacity distribution that is absolutely better than the others, but it is dependent on the thermo-physical characteristics of the internal elements. The optimized external wall stratigraphy can be composed of an even number of resistive and capacitive layers, alternating with each other, or of an odd number of layers (with one more resistive layer than the capacitive ones). Even the number of layers that makes optimized the passive behaviour of the external wall varies with the characteristics of the internal elements, the results show that already with a number of layers varying from 2 to 5 high performances can be obtained.
Read full abstract