Search

Summary: This paper concerns the rheological characterisation of liquefied sands as non‐Newtonian Bingham fluids. For this purpose, dam‐breaking laboratory tests are often executed and interpreted, offering a viable option to identify the properties of fluidised water‐soil mixtures. However, limited attention has been devoted so far to clarify what variables and measurements would allow unambiguous calibration of Bingham parameters, namely, the viscosity η and the yield stress τy. The numerical results of parametric studies based on the particle finite element method (PFEM) are critically inspected to gain deeper insight into the problem. First, it is confirmed that multiple η − τy pairs may reproduce the same experimental evidence when formed by only one measurement—usually, the post–dam‐breaking displacement of the bottom toe (tip) of the liquefied mass. Then, two alternative procedures are proposed for unambiguous identification of both η and τy: one is based on monitoring the evolving aspect ratio of the fluid mass during free, gravity‐driven flow; the other relies on a slightly different dam‐breaking test, also including impact against a rigid obstacle. In particular, the latter approach reduces the relevant duration of the test, reducing the possible influence of reconsolidation effects on the calibration of rheological parameters. [ABSTRACT FROM AUTHOR]

SUMMARY A Lagrangian numerical approach for the simulation of rapid landslide runouts is presented and discussed. The simulation approach is based on the so-called Particle Finite Element Method. The moving soil mass is assumed to obey a rigid-viscoplastic, non-dilatant Drucker-Prager constitutive law, which is cast in the form of a regularized, pressure-sensitive Bingham model. Unlike in classical formulations of computational fluid mechanics, where no-slip boundary conditions are assumed, basal slip boundary conditions are introduced to account for the specific nature of the landslide-basal surface interface. The basal slip conditions are formulated in the form of modified Navier boundary conditions, with a pressure-sensitive threshold. A special mixed Eulerian-Lagrangian formulation is used for the elements on the basal interface to accommodate the new slip conditions into the Particle Finite Element Method framework. To avoid inconsistencies in the presence of complex shapes of the basal surface, the no-flux condition through the basal surface is relaxed using a penalty approach. The proposed model is validated by simulating both laboratory tests and a real large-scale problem, and the critical role of the basal slip is elucidated. Copyright © 2016 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]