x
Diese Website verwendet Cookies. Mit der Nutzung der Website stimmen Sie deren Verwendung zu. Weitere Informationen erhalten Sie in unserer Datenschutzerklärung.

Non-Linear Fracture Mechanics in LS-DYNA and LS-PrePost

In non-linear fracture mechanics, an energy based criterion is used for assessing the risk for crack growth: if the energy release rate at the crack tip exceeds what is required for creating new surfaces in the material, crack growth will occur. Under certain assumptions the energy release rate at the crack tip can be calculated by a path independent integral, the so-called J-integral. In modern FE-based fracture mechanics applied to practical design, the structure under consideration is modelled, including cracks at specific locations, and the J-integral values are computed and used as design criteria. From a numerics viewpoint, the J-integral has many appealing properties: it can be evaluated from the far-field solution, which reduces numerical errors that may arise close to the crack tip, and the expected path-independence can to some extent be used as a quick check on solution validity. Evaluation of the J-integral from LS-DYNA simulation results has been implemented as a post-processing tool in LS-PrePost, including consistent treatment of residual stresses. The implementation covers both 2D (plane stress / plane strain) and 3D applications, using the virtual crack-tip extension (VCE) method. The tool is accessible both via the LS-PrePost GUI and via command file interface. The J-integral evaluation capability opens for new possibilities to use LS DYNA within fields where classical fracture mechanics is applied for assessment of structural integrity, for example the nuclear industry, the offshore industry, and general pressure vessel design. Combining the previously presented procedure for welding simulations (*MAT_CWM) with modelling of cracks in the weld area and the new J-integral tool of LS-PrePost makes it possible to evaluate the effect of cracks on the structural integrity of the welded structure, including also the effects of the welding process. Results for some benchmark examples and comparisons to other FE-codes will be presented, demonstrating the validity of the implementation.