1. Introduction and warning

To study the degradation of a structure, several tools are available and in particular (by increasing level of complexity and calculation time):

• the mechanics of rupture with triggering criteria based on stress criteria or the energy return rate;

• cohesive zone models that make it possible to study the initiation and propagation of cracks when the potential crack path is known a prima facie;

• damage models that make it possible to model the degradation of the material in a diffuse or localized way.

Implementation guides in Code_Aster exist for the first 2 families of methods. These are documents U2.05.01 and U2.05.07. This document aims to complete the range of user guides by focusing on the implementation of damage calculations in a near static manner.

However, it should be noted that, compared to other techniques, damage calculations are the most complicated to implement, the most expensive in terms of calculation time but also those that are the least controlled both from the point of view of robustness and from the point of view of the quality of the result, in particular because:

• damage problems generally do not have a single solution at the structural level;

• the solutions obtained with local damage models are highly dependent on the mesh;

• it is not yet possible to propose ways of identifying the characteristic lengths of materials, and these lengths cannot be transposed from one regularization method to another;

• the management of broken meshes is often a problem, especially with regularized models. According to the models, we can observe non-physical widenings of the damaged bands rather than the appearance of multi-cracking or very strong convergence difficulties.

It is therefore important to always take a critical look at the results obtained.