A Comparison of Two Elastic-Plastic Approaches To Cyclic Soil Modeling

Abstract eng:
The choice of the soil constitutive relationship is an important step in the numerical modelling of many Geotechnical Earthquake Engineering problems. Although relevant results have been achieved in the field of elastic-plastic constitutive soil modelling, many practical problems are still tackled through 1D visco-elastic models, only based on shear modulus reduction and material damping curves. This is partly motivated by the usual lack of experimental data and the related difficulties in the calibration of advanced soil models. Even though the number of material parameters is kept low, multiple options exist in terms of theoretical framework for the model formulation. In this paper, the cyclic performances of two different kinematic hardening frictional models are compared. Both the models are characterized by a Drucker-Prager-type pressure-sensitiveness and non-associativeness, but, while the former is formulated as a standard elastic-plastic model with Armstrong-Frederick rotational hardening, the latter has been recently developed in the framework of bounding surface plasticity with vanishing elastic region. In particular, the bounding surface model allows for soil plastification at any load level, extending to frictional materials the previous cohesive version by Borja and Amies (1994). Although the two models are characterized by similar analytical structures and the same number of material parameters, the cyclic performance of the bounding surface model is apparently the most satisfactory in terms of simulated stiffness degradation and damping curves, which can be readily exploited for the calibration of the hardening parameters. The dissipation properties of both models are investigated in conjunction with an additional viscous mechanism, exploitable to improve the simulation of the experimental damping. The effectiveness the bounding surface model is further analysed against the influence of the initial confining pressure. In particular, it is shown that the simple introduction of pressuredependent hardening parameters readily allows to reproduce the soil cyclic response over a wide range of confining pressures.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket