Frost Damage Risk in Insulated Masonry Walls: Challenges in Analysing Large Number of Numerical Simulations

Abstract

This paper presents and applies a methodology for conducting and analysing large number of numerical simulations to investigate the risk of frost damage in internally insulated masonry walls. In this context, 2500 simulations were conducted with Delphin 6 software following a full factorial design with six input variables: insulation material, insulation thickness, brick thickness, brick type, interior water vapour production and wall orientation. The methodology is applied to investigate four different aspects: (1) the choice of damage indicator, (2) the impact of critical temperature and moisture content, (3) the location of investigation and (4) the impact of wall properties and boundary conditions on the risk of frost damage. The simulations show a good correlation between the different indicators and point out the large impact of critical values for temperature and moisture content on the risk of frost damage. It also shows that brick type, insulation thickness and orientation are important parameters when looking at the risk of frost damage. Lastly, it suggests that an investigation point at 10 mm from the exterior surface should be preferred rather than the 5 mm generally used in the literature. The main limitation of this study is related to its transferability. On the one hand, transferring the methodology would require a simple adaptation of the statistical model used. On the other hand, transferring the results obtained from applying the methodology is more complicated since those are highly case-related. Further work should focus on the generalization of these conclusions, including – but not limited to – other climatic data and other material properties.