Validation of the Computational Model

International audience; This chapter compares uniaxial experiments and tests on hollow cylinders to computations. It evaluates the capacity of the computational model to capture the main features of the experiments, that is the growth of permeability with increasing electrical energy inserted into the system. The model calibration has been carried out using a small set of experiments and comparisons with further simulations on the rest of the data provided some evaluation of the quality of the finite element model. The most demonstrative and quantitative evaluation has been provided by the evolution of permeability with increasing electrical energy. In most approaches, permeability is isotropic and it is calculated according to the maximum principal damage. The average permeability of the specimen is overestimated in the isotropic case compared to the anisotropic case. Losing the effect of the directionality of damage on the permeability yields a severe overestimation of the overall permeability.