SUMMARYThis paper describes a method of performing the integration of generalized plasticity models, in which, unlike classical elastoplasticity, the yield surface is not explicitly defined. The algorithm is based on a generalized midpoint scheme and is applied to a specific generalized plasticity model for sands, in which a hyperelastic formulation is introduced to describe the reversible component of the soil response instead of the hypoelastic approach originally proposed. In this way, an efficient integration scheme is developed in the elastic strain space. The consistent, algorithmic tangent operator is derived. Isoerror maps are generated to study the local accuracy of the numerical integration algorithm. Results from a series of numerical examples based on the simulation of drained triaxial tests are given to illustrate the accuracy and convergence properties of the algorithm, both at the local and at the global level. Finally an example is given of the simulation of a cyclic triaxial test to illustrate the improvement on accuracy caused by the use of a hyperelastic law into the constitutive equations, as opposed to the hypoelastic formulation initially adopted in the model.
In this paper, the mechanical response of silty sediments forming the upper profile of the Venice lagoon basin is modelled within a theoretical framework known as Generalized Plasticity. Starting from an existing formulation developed for the analysis of sand behaviour, some modifications are introduced in the original model in order to better simulate the stress-strain response of such non-active natural soils over a wide range of stresses and densities. A state dependent dilatancy is included in the constitutive equations. Moreover, according to recent developments on the isotropic compression of granular soils and on the modelling of softening in dense sands, adjustments to the plastic modulus expression are introduced. The approach is validated by comparing the model predictions with experimental results of drained triaxial compression tests on natural and reconstituted samples of soils, having different fine contents.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.