Cerfacs Enter the world of high performance ...

Simulations of COVID propagation

 , , ,

Required Education : Master or engineering schools
Start date : 2 February 2021
Mission duration : 36 months
Deadline for applications : 26 February 2021
Salary : 2475 Euros brut/month

There is no doubt that the COVID epidemy raises major questions for our society. To control the epidemy, a central question is the famous Ro factor: how many persons get infected by one sick person.


This question depends on biological and medical factors but it is also controlled strongly by fluid mechanics: the virus is carried by small water droplets which evaporate and are dispersed in the atmosphere. They can move in a room and remain in suspension for long times. They can also stick to walls and to human skin. The objective of the PhD is to simulate these phenomena using the tools developed for the aerospace industry at CERFACS to simulate kerosene spray injection in engines.

The simulation tool will be a direct evolution of the AVBP code which is the world reference for reacting flows. Its Lagrangian solver will be adapted to the specificities of COVID in collaboration with doctors and laboratories specialized in virology. The solver will also be adapted to compute long physical times: while the flow in a combustion chamber is established within a few seconds, it takes hours of physical time before the virus distribution reaches steady state in an aircraft for example.

This multidisciplinary thesis will be performed in collaboration with doctors, companies and government agencies to provide a high-fidelity simulation tool which can provide risk evaluation with precision. For example, the risk raised by having one sick person breathing in an aircraft cabin or in a car for hours will be an output of the tool. Hundreds of possible scenarii will be simulated in parallel to account for different A/C tunings, sizes, positions of passengers, possible mitigation strategies etc.

Candidates must have a strong background in fluid mechanics and High Performance Computing. No knowledge on medicine is needed but the motivation to interact with different scientists is mandatory.

Candidates will be PhD students registered in Université de Toulouse for 3 years. The salary corresponds to the standard conditions at CERFACS (which are quite good…).



T. Poinsot poinsot@cerfacs.fr

F. Duchaine duchaine@cerfacs.fr

N. Odier odier@cerfacs.fr