Analysis of the propagation

Leader: Prof. Claudio di Prisco (Politecnico di Milano)

Dry theoretical model 

Why? Granular systems can be treated as solids if the particles are densely packed and a network of contacts develops within the medium, or like fluids whenever the grains are largely spaced and free to move in any direction, interacting only through collisions. As a consequence, a key question concerns how to model the transition from fluid to solid-like behavior and vice-versa when describing the mechanics of a granular material from a continuum point of view.

 

What? Constitutive model capable of capturing the response of dry granular flows from quasi-static to dynamic conditions, in particular when the material experiences a sort of solid-to-fluid phase transition.

How? In the constitutive model, both concentration and granular temperature are chosen as state variables, and the stress tensor is computed as the sum of two contributions: the quasi-static and the collisional one. The former one is determined by using an elasto-plastic model, while the latter one is derived from the kinetic theory of granular gases.

Preliminary results
        Steady state:
theoretical model compared with DEM true triaxial simulations
        (Redaelli & di Prisco, 2018, submitted)

        Evolving conditions: theoretical model compared with DEM constant volume shear tests
        (Vescovi et al., 2019, in preparation)

Saturated theoretical model

Why? Rapid flow-like landslides are triggered by heavy rainfall, as a consequence water plays an important role in both the triggering (the most critical triggering factor being a hydraulic condition such as either rain infiltration or water table rising) and the propagation phase, governed this latter by the hydro-mechanical coupling. From a micro-mechanical point of view, the presence of water in saturated soils changes the dynamics of the grain-grain interactions, and produces additional dissipations due to the grain-water contacts.

What? Advanced version of the dry constitutive model to consider liquid within pores and able to account for granular-liquid coupling effects.

How? The model is formulated within the context of the two-phase mixture theory and, in order to account for the dissipative mechanisms of grain-grain and grain-water interaction, the energy balance equation for the two phases is used. The extension of the dry model to saturated conditions is achieved by: (i) considering the additional balance equations for the liquid phase, (ii) including additional dissipative contributions accounting for the presence of water, and (iii) adopting a suitable definition for the liquid viscosity depending on the granular concentration.

Preliminary results
        Steady state: theoretical model compared with DEM constant volume shear tests.
        (Marveggio Master thesis, 2018; Vescovi et al., 2019, submitted)

Numerical simulations of rapid landslides along natural slopes

Why? Simulating the landslide propagation is fundamental for a reliable risk-assessment. These phenomena can evolve in the form of short runout sliding or as long runout flows with an extreme erosive power. A suitable numerical tool has to be capable of dealing with initial change in evolution, large displacements and strain rates, multiphase granular materials governed by hydro- mechanical processes and soil-structure interaction.

From a numerical point of view, the increasing success of computational tools conceived to deal with large displacements has made possible the quantitative numerical simulation of granular flows.

According to the literature, continuum-based methods proved to be able to deal with small and large deformations of soils using a macroscopic description of the material constitutive behavior. Smoothed Particle Hydrodynamics (SPH), Particle Finite Element Method (PFEM), and Material Point Method (MPM) proved to be valuable options.

While the theoretical background can be widely different, these methods share a set of beneficial features. They are (i) suitable for complex geometrical boundaries; (ii) free from fixed discretization meshes; (iii) easily parallelizable. At the moment, it remains unclear which is the best numerical tool to be employed in terms of accuracy and efficiency.

What? Quantitative parametric analysis of the propagation processes by employing advanced numerical models (SPH, PFEM and MPM) and suitable material rheologies.

How? The 3 numerical methods are used to perform simulations of rapid landslides. The analyzes are carried out considering a mass of unstable soil, described by a Bingham model, positioned on top of a slope. The influence of the geometry of the slope and the volume of the soil mass on the runout travelled by the soil mass and its final geometry are investigated.

Preliminary results
        Comparison of the three methods in terms of reliability of the predictions and computational costs.
        (Rovelli Master thesis 2018; Gaudio et al. 2018)

On going research

  • Calibration of the model parameters on the base of DEM numerical results
  • Implementation of advanced constitutive relations for evolving 3D conditions
  • Numerical implementation of the proposed model in a MPM code

Researchers involved in this project

Claudio Giulio di Prisco, Pietro Marveggio, Irene Redaelli, Dalila Vescovi