COST Action FP0802Experimental and Computational Micro-Characterisation Techniques in Wood Mechanics

WG3: Computational modelling

In this Working Group, computational models for the hygro-mechanical behaviour of wood will be developed. As in Working Group 2, elasticity, viscoelasticity, viscoplasticity, mechano-sorption, hygroexpansion, moisture transport and sorption will be considered. Research efforts will focus on further development of existing models, which are mostly limited to particular physical effects and specific length scales, and on their combination and integration. Appropriate physically-based analytical descriptions for the various couplings, resulting mainly from the hygroscopicity of the material, and their appropriate implementation into computational schemes will constitute the greatest challenge in this Working Group. The applied computational techniques are summarised in the list below. Emphasis will be placed on the experimental validation of the models. Test results obtained in the context of Working Groups 1 and 2 will be used for this purpose. Moreover, test data from these areas will also be employed for the identification of the model input parameters.

The computer simulations will contribute to an enhanced understanding of the relations between microstructural features and properties and will allow studying phenomena and identifying properties not accessible to investigations by experimental techniques. Moreover, the developed simulation tools will also be used for "virtual testing" and "reverse identification". Virtual testing denotes the numerical simulation of a test on the computer. Prior to a test, it allows for investigation of various test configurations and, thus, for optimisation of the test design. After the test, the influence of the test conditions (e.g. relative humidity) can be analysed and corresponding results that would have been obtained under other conditions can be estimated. This contributes to increased comparability of test results. Reverse identification refers to the back-calculation of material properties from test results. For example, numerical simulation of a nanoindentation test allows for determination of material properties of the sample from measured data pairs of applied forces and corresponding indentation depths. Obviously, the suitability of the identified material parameters depends strongly on the suitability of the applied computer models. This underlines the importance of great efforts in the model development and a thorough experimental validation.

Close Window