5. The field of validity#

Several checks are possible to verify that the law of behavior chosen, and the values of the parameters used, are valid for the simulation.

In addition to the advice given above, to properly choose the law of behavior according to what you want to model, some additional checks can be carried out using specific tools.

5.1. Validity of the parameters in the deformation and speed range.#

Since the parameters of the chosen model are identified within a certain deformation range, it is important to verify that in studies using these parameters, these deformations remain well within the identification interval.

The traction curves defined by DEFI_FONCTION incorporate a « safeguard »: the maximum value of the abscissa (EPSI) cannot be exceeded in the study. But if this ever happens, instead of defining a constant (or worse: linear) extension, it is appropriate to resume the identification to define additional points in the traction curve.

Linear work hardenings (ECRO_LINE), or those defined by an analytical function (ECRO_PUIS, VMIS_CINx *_ CHAB, etc.) are much more dangerous. Nothing will prevent studies from greatly exceeding the level of identification distortion. This is why protection should be implemented in a future release.

In any case, it is relatively easy, in post-processing a study, to calculate (CALC_CHAMP) the norm of the deformation field (EPEQ_ELGA) and to extract the maximum (POST_ELEM/MINMAX, or graphic post-processing in SALOME_MECA).

If the study leads to the use of a formalism of large deformations, it is necessary for identification to use it as well.

As far as the deformation rate is concerned, again a check is necessary. Its automatic calculation should be proposed in a future version.

5.2. Discharge: validity of isotropic work hardening (and Hencky’s laws)#

How to verify that the landfills are small enough for the calculation with isotropic work hardening to be valid? In CALC_CHAMP there is a DERA_ELGA discharge indicator (cf. u4.81.04 u4.81.04).

  • The components DCHA_V, DCHA_T indicate whether there are unloads on the constraints (either on Von Mises or on the total tensor), thus invalidating the calculation with a nonlinear elastic law.

  • Component IND_DCHA provides an indicator that indicates whether there is a risk of re-entering plasticity in landfill, thus invalidating the isotropic work hardening calculation.

For more details on how to calculate them, see r4.20.01 Indicators of discharge and loss of proportionality of elastoplasticity loading).

5.3. Radiality: effects of non-proportionality#

In the case of highly non-proportional cyclic loads, the effect of over-work hardening can be ignored by the chosen behavior. By using the discharge and radiality of the load indicator in CALC_CHAMP: DERA_ELGA (cf. u4.81.04), the ERR_RADI component measures the error made by the rotation from normal to the load surface. If this value is important, you must then use a model to take this effect into account (for example VISCOCHAB).