Computer aided identification of the moisture transport parameters in spruce wood

An inverse identification method for characterization of wood sorptive properties is presented. The method relies on a computer simulation of a real experiment, in our case a desorption experiment, where spruce heartwood samples were dried from 27 % to 8 % moisture content. Three samples, distinguished by the respective moisture flow pattern through the specimen, were investigated. A computer aided material characterization using the so-called inverse problem identification method was performed on the measurements. The solution of the specified inverse problem enabled us to estimate the moisture diffusion coefficients of wood and to determine the moisture content field in the sample simultaneously. The method is first verified on two simple cases of uniaxial moisture flow, and then is used to characterize the diffusion coefficients on a biaxial moisture flow sample. In the latter case some salient features of the proposed method are exhibited.