Focus on one of InPhyNi's research projects benefiting from access to the Azzurra cluster
Concentrated suspensions of solid particles are ubiquitous in industry as well as in biological and natural flows: fresh concrete, launcher propellants, blood and lava are just a few examples.
The rheological properties of these suspensions are complex and still partly misunderstood. They result in particular from the cross-effects of hydrodynamic interactions and contact between particles. Due to the complexity of the mechanisms at play at the particle scale, numerical simulations are a privileged tool for exploring the connection between microscopic properties and macroscopic behaviors. In addition, some of the open questions concern the transport of particles (“shear-induced migration"), the study of which requires the simulation of large systems.
We are developing a massively parallelized numerical method for the three-dimensional simulation of such dense, polydisperse, non-Brownian suspensions. This method is based on a fictitious domain approach, where the fluid and solid particles share the same mesh. The rigidity of the particles is ensured by the computation of a suitable force field inside the particles.
We have implemented this method in the OpenFOAM toolbox, a free and open source toolbox for solving fluid mechanics equations using finite volume methods. We intend to propose a complete model that takes into account all the phenomena characterizing these suspensions at the particle scale - long-range hydrodynamic interactions, lubrication between particles and with the walls, contact forces. We will thus be able to study in detail the influence of these mechanisms on the rheology, to better understand the experimental measurements and finally to propose continuum rheological and particle transport models, which can be used for the computation of complex flows.
The Azzurra HPC center - the new HPC cluster at UCA - provides us with the necessary computing resources. We benefit from both high-performance hardware and efficient technical support, enabling us to make rapid progress. Therefore, Azzurra is an essential resource for our project.
This project is supported by the ANR (AMARheo project ANR-18-CE06-0009).
