SUMMARYThis paper presents a constitutive model for unsaturated soils with only 13 parameters. Its development is based on the concept of bounding surface plasticity under a critical state framework. The definition of effective stress via an equivalent pore pressure and the introduction of suction effects into a general plastic hardening mechanism are inspired from the work of Pereira et al. (Ph.D. Thesis, 2005) and that of a few others. Despite the minimal number of parameters, this model can simulate correctly complex volumetric behaviour and post-peak softening, as well as a few other classic typical behaviours of unsaturated soils. The model is validated using experimental data both on saturated and unsaturated soils. Its efficiency has been highlighted by comparing its outputs with those from the Barcelona Basic Model and from another advanced model CJS-NS. The numerical studies revealed the necessity of additional experimental data for a more complete validation of the model and its future improvement.

Summary This article focuses on modeling the strain hardening‐softening response of statically compacted silty sand as observed from a comprehensive series of suction‐controlled, consolidated‐drained triaxial tests accomplished in a fully automated, double‐walled triaxial test system via the axis‐translation technique. The constitutive model used in this work is based on the theory of Bounding Surface (BS) plasticity and is formulated within a critical state framework. The essential BS model parameters are calibrated using the full set of triaxial test results and then used for predictions of compacted silty sand response at matric suction states varying from 50 to 750 kPa. Complementary simulations using the Barcelona Basic Model have also been included, alongside BS model predictions, in order to get further enlightening insights into some of the main limitations and challenges facing both frameworks within the context of the experimental evidence resulting from the present research effort. In general, irrespective of the value of matric suction applied, the Barcelona Basic Model performs relatively well in predicting response at peak and critical state failure under low net confining pressure while the Bounding Surface Model performs relatively well under high net confining pressures.

No abstract