Direct numerical simulation for magnetic flow control

Post-doc project

Flow control involves passive or active devices designed to obtain a beneficial effect on the flow for industrial purposes. Under certain conditions, the application of a constant (DC) magnetic field to an electro-conductive liquid creates a net body force into the flow that can be used to manipulate the flow in a certain way to obtain stabilisation, drag reduction, heat transfer or mixing enhancement, stirring… Depending on the targeted application, the principle can apply to strongly electro-conductive media (i.e. liquid metal) or moderately electro conductive media (i.e. standard electrolytes, molten salt…)

Such technique of flow control by externally applied magnetic field involves the mutual interaction of the velocity field of the fluid and the electro-magnetic field (Magneto-hydro-dynamics, MHD).

Direct numerical simulation of a flow control situation where two different systems are used: a uniform magnetic field is used (video at the top); a non uniform magnetic field generated by magnets placed around the walls of the channel (video in the middle).

The main objective of the project is to develop a MHD resolution tool (without modelling approximation) via a direct numerical simulation methodology. This work aims at finely quantifying the modification experienced by the flow under a constant (i.e. DC) but spatially evolving magnetic force.


A particular attention is paid to the integration of the module into the existing multiphysics direct numerical simulation code, in order to address a large range of industrial and academic scientific challenges in a very near future.


As an application case, different mapping of wall magnets distribution are studied, and the induced flow modification and their effects on the associated heat transfer for highly (liquid metals) and moderately conductive media are assessed. Valuable information about the efficiency control parameters (i.e. magnets localisation and sizes…) is expected as a result of this project, and should be very helpful for designing or refining flow control strategies.


  • PI: Olivier Doche
  • Co-PI: Sedat Tardu
  • Post-doc: Jonathan Schillings


  • LEGI