Influence of uncertainty in heat-moisture transport properties on convective drying of porous materials by numerical modelling

The influence of uncertainties in heat–moisture transport properties, due to measurement errors and material heterogeneity, on the numerical simulation results of convective drying of two capillary-saturated porous materials is investigated by a transport-property parameter analysis (TP-PA), based on the Monte Carlo method. Here, the heat–air–moisture transfer model is evaluated many times, each time using a random set of one or multiple input parameters (i.e. transport properties), by which a stochastic model output is generated. The propagation of these transport-property uncertainties to the hygrothermal behaviour of the drying system is evaluated by statistical analysis of the model simulation output. The spread on the hygrothermal response is found to be strongly material dependent and is related to the dominant mode of moisture transport in the material, i.e. liquid or vapour transport. The TP-PA results clearly indicate that uncertainties in the heat–moisture transport properties can lead to significant differences in drying behaviour predictions, where differences of the total drying time with respect to its mean value up to 200% are found for the materials considered. Therefore, numerical modelling of heat and moisture transport in porous materials should preferably include a quantification of the propagation of these uncertainties, for example by means of the proposed transport-property parameter analysis. Such analysis, however, additionally requires detailed (a priori) experimental material characterisation to determine realistic uncertainty ranges.