To protect a flat roof from interstitial condensation, a vapour retarder is required in cold climates. However, in the case of an autoclaved aerated concrete (AAC) roof a vapour retarder would also prevent the construction moisture from drying, and thereby, severely impair the thermal insulation quality of the AAC. In order to find out whether an unvented flat roof of AAC works without a vapour retarder field test or computational simulations can be made. Experimental investigations are expensive and of limited transferability. An alternative is the use of validated models to assess the hygrothermal behaviour of proposed constructions. Until now the uncertainty of input data was explicitly left out of hygrothermal modelling. This was done because the understanding of the individual physical processes and their impact on the component assembly was the first priority. In the following work the necessary input data for hygrothermal calculations are described with a specific uncertainty. Today, an increasing demand exists to define more realistically processes, which also include kind and dimension of the element of uncertainty [1, 2, 3].

2. Numerical Approach

This work is focused on two uncertainty approaches for hygrothermal building simulations; the sensitivity analysis and the probability (stochastically) based analysis. With the help of the sensitivity analysis, one can study the sensitivity of a problem solution based on the data confidence input and its reaction to a single parameter of uncertainty. If information about the stochastically deviation of the influencing factors is known, this can be included in a probability based analysis. Both types of analysis of the input data are included in the computer program WUFI, which allows the calculation of the transient heat and moisture, transport in building assemblies [4, 5] and was developed at the Fraunhofer Institute for Building Physics in Holzkirchen. WUFI has repeatedly been validated by comparison with experimental data. The drying behaviour of an AAC roof under different exterior and interior climate conditions is object of this study.

• HAM: computer models : WUFI
• mathematics: error analysis, uncertainty analysis
• mathematics: sensitivity analysis, parametrical variation : climatic data
• mathematics: sensitivity analysis, parametrical variation : material properties
• mathematics: stochastical modeling and analysis
• roof
• weather, climates : sensitivity to climatic data on moisture balance

1. An educational hygrothermal model: WUFI-ORNL/IBP
2. Combined effect of temperature and humidity of the detoriation process of insulation materials in ETICS
3. Determination of moisture and salt content distributions by combining NMR and gamma ray measurements
4. Moisture buffering effects of interior linings made from wood or wood based products
5. Moisture-buffering effect - experimental investigations and validation
6. Non-isothermal moisture transfer in porous building materials
7. Position paper on material characterization and HAM model benchmarking
8. Practical application of an uncertainty approach for hygrothermal building simulations--drying of an AAC flat roof
9. Previous Experimental Studies and Field Measurements on Moisture Buffering by Indoor Surface Materials
10. Simulation of indoor temperature and humidity conditions including hygrothermal interactions with the building envelope
11. Stochastic building envelope modeling -- the influence of material properties
12. The hygrothermal behaviour of rooms: combining thermal building simulation and hygrothermal envelope calculation
13. Two-dimensional transient heat and moisture simulations of rising damp with WUFI 2D
14. Uncertainty approaches for hygrothermal building simulations - drying of an AAC flat roof in different climates
15. Uncertainty of hygrothermal calculations