An XFEM-approach to model brittle failure of wood. (1st June 2020)

• Record Type: Journal Article
• Authors: Gebhardt, Clemens; Kaliske, Michael
• Is Part Of: Engineering structures. Volume 212(2020)
• Publication Date: 2020-06-01
• Subjects: Wood -- Timber -- Fracture -- XFEM -- Strength-grading -- Numerical simulation -- Cohesive zones; Structural engineering -- Periodicals; Structural analysis (Engineering) -- Periodicals; Construction, Technique de la -- Périodiques; Génie parasismique -- Périodiques; Pression du vent -- Périodiques; Earthquake engineering; Structural engineering; Wind-pressure; Periodicals; 624.105
• DOI: 10.1016/j.engstruct.2020.110236
• Languages: English
• ISSNs: 0141-0296

Abstract

Highlights: Adaptive cohesive crack growth model for wood using XFEM and a multi‐surface‐failure criterion based on stress field. Coupling of cracks, material interfaces and crack branches in a regular mesh using XFEM. Presentation of a crack growth strategy for inhomogeneous wooden boards with ingrown knots. Validation of the method with experiments of common wooden structures and boards with knots. Application to strength grading of wooden boards. Abstract: A comprehensive approach to model the brittle failure of arbitrary wooden structures is presented. The Extended Finite Element Method (XFEM) is utilised to depict discontinuities in the displacement field, i.e. cracks and to model material interfaces in the strain field. Location and direction of crack nucleation and crack growth are identified by a stress based failure criterion utilising a multi-surface formulation. An anisotropic cohesive formulation is used to depict the fracture process zone. A new approach for the simulation of wooden boards with inhomogeneities is presented, which models the ingrown knots with a coupled weak and strong interface. A completely automatic simulation of crack growth on a regular mesh modelling the material and cohesive interface between knot and surrounding material at the same time is feasible. The proposed methods are validated by comparison to experiments.