Paraelastic description of small-strain behaviour
Author: Prada Sarmiento, Luis Felipe
Director(s)/Advisor(s): Lizcano Peláez, Arcesio; Niemunis, Andrzej; Meziat Vélez, René Joaquín; Rodríguez Herrera, Carlos Francisco
Publication date: 2011
Content type: doctoralThesis
The adequate characterization of the small-strain stiffness soil behavior plays an important role in the prediction of displacements of retaining walls and footings subjected to monotonic and/or cyclic loading. This document recounts the main experimental features observed on monotonic and cyclic element tests performed on sandy and clayey soil samples that allowed to coin the term small-strain stiffness in the 1970?s. The starting point in this thesis is the analysis of the hypoplastic equation with intergranular strain. The basic version of hypoplasticity cannot reproduce many aspects of the cyclic behavior observed on sands. The concept of intergranular strain (IS) was the introduced as an internal variable needed to memorise the most recent strain history in order to reduce the ratcheting of the model and to take into account abrupt changes of stiffness upon changes of straining direction. However, a significant accumulation of strain is still obtained within the small-strain range (SSR). Some parameters of the IS must be adjusted to the amplitude of the strain loops and even them may lead to overshooting problems. Kinks in the damping ratio curves are obtained for strain amplitudes comprised in the SSR. Elastoplastic models based on the bounding surface plasticity concept perform well when simulating cyclic loading, specially for large strain amplitudes. Models like Sanisand define a very narrow yield surface under the concept of vanishing elastic region. Within the Ducker-Prager cone, a linear hypoelasticity controls the soil response, rendering the description of the small-strain stiffness effects inaccurate. No damping can be observed for yampl< 10-4. The inaccurate description of some features of the small-strain stiffness behavior observed both in hypoplasticity with intergranular strain and elastoplastic models can be repaired with the reformulated version of Hueckel & Nova?s paraelasticity. The new paraelasticity has been devised as a non-linear elastic model comprised in the range of small strains (yampl< 10-4). Whithin this range the perfect reversibility of stress and strain cycles is guaranteed at all reversal points (also in full 3D case). The model follows the masing rules and reproduces the degradation of the shear modulus and the?
Files in this item