Dense gravitational flows

The aim of this lab-course is to tackle the problem of the modeling of dense gravitational flows dynamics and the mitigation of avalanches.

Dense flows of granular materials will be produced and analyzed with the help of two laboratory devices: a large inclined plane equipped with advanced instrumentation (granular PIV, fringe projection, etc.) and a reduced model with similar yet simplified instrumentation. Emphasis will be placed on the problem of abrupt variations of flow depths, velocity and density, namely granular jumps, that occur when granular flows impact walls. The laboratory tests will be backed up with theory and numerical simulations, with the objective of inferring the relevant rheological parameters of the studied granular fluid.


Experimental study of a model granular flow down an incline: analysis of the behaviour using dedicated measurement techniques

4 hrs session

In this first part, an experimental device is used to generate a gravitational flow with calibrated glass beads serving as a model granular material. The phenomenon is observed with a number of different techniques such as ultrasonic flow height sensors and high speed video cameras enabling to assess the thickness of the flow, its shape and velocity (using Particle Image Velocimetry, PIV).

By varying key parameters such as the initial mass or volume of the granular material, its composition and the slope angle of the chute zone and the run-out zone, the participants analyse the behaviour of the flow and put it in perspective with the shallow-flow model of Saint-Venant.

Numerical simulation of the granular flow: assessment of the key parameters involved in the propagation using the "shallow-flow" model

4 hrs session

In order to refine their understanding of the phenomenon, the participants carry out numerical experiments based on the Saint-Venant model. After fixing the main parameters (total volume of the avalanche, flow height at the upstream boundary conditions, and the friction), a numerical simulation is launched and the results are compared with the observed experimental data.

Some parameters are numerically adjusted in order to obtain an agreement between the model and the experiments, and infer the relevant rheological parameter of the studied granular material.



INRAE Grenoble
2 rue de la papeterie