SIMQUA : un logiciel de simulation de la qualité du bois

À partir des données morphologiques d'une grume (forme extérieure, nature de la branchaison, taille des accroissements annuels) et des caractéristiques physiques et macroscopiques du matériau telles que la densité du bois et l'angle du fil (fibre torse), nous avons développé un logiciel qui permet la simulation d'un débit scié dont nous pouvons décrire la qualité d'aspect des 4 faces et calculer sa masse volumique. Ce logiciel ouvre des perspectives intéressantes pour l'évaluation objective de l'influence des conditions de croissance d'un arbre ou d'un peuplement sur la qualité du bois produit.Modelling wood quality. New software : SIMQUA. Forest management requires an understanding of the effect that silvicultural treatments have on wood properties. The forest manager must know the wood quality produced by different silvicultural treatments in order to be able to choose the right one. A variety of software models the effects of cultural practices and environmental factors on the growth and yield of trees (eg Douglas fir); we decided to create software for modelling wood quality with a view to linking it with different growth models. In this work the following aspects of wood quality have been selected: size and distribution of the branches (which are described by whorl, each branch being alive, dead or pruned), grain angle, ring width and pith position at different levels in the tree. The specific gravity can be defined for each ring. According to the model developed the log can be defined with extreme accuracy. The mathematical functions used are the Lagrange polynomial of degree n-1 (useful in the finite element method). It is now possible to simulate the sawing of boards from the described log and to compute the position, size and shape of knots on the 4 faces of each board and the specific gravity of each board. These results permit us to simulate different visual rules to estimate the quality of the sawn logs. In future it will be possible to optimize the sawing by visual or mechanical criteria. The data needed by this software (ring width, pith position, description of the branches) can be measured or produced by a growth model. The models is able to predict specific gravity variation in the tree and the angle and diameter of the branches in the different whorls must be produced before our new tool totally efficient