Original Experimental and Numerical Approach for Prediction of the Microscopic Hygro-Mechanical Behavior of Spruce Wood

Being a hygroscopic material, wood is subjected to dimensional variation from a hygric state to another. These variations affect strongly its physical properties and consequently its structural functions in the building. The quantification of the swelling during the hygric transfer within the wood elements is still unknown by the researchers. Hence, specimens with representative dimensions have been prepared and conditioned at two different relative humidity conditions. At each humidity state, the specimens have been scanned using X-ray micro-tomography to have access to their real morphology and evaluate their swelling. After that, a global approach has been set in order to model the hygro-mechanical behavior of the material and validate the model based on the experimental results. To do so, the displacement fields of the volume from the dry to wet humidity states have been calculated with Digital Volume Correlation (DVC). Moreover, a microscopic hygro-mechanical model on the real 3D structure of the material has been proposed. The model has been further validated, through an identification approach, in order to minimize the gap between the numerical and the experimental displacements. The results show good convergence of the model to the experiments, and an estimation of the Poisson coefficient and swelling coefficient of the cell walls. (C) 2019 Elsevier Ltd. All rights reserved.