Modeling and Information Processing

Applied mathematics and scientific computing

The issues

The numerical simulation of physical phenomena figures prominently in all phases of the design and evaluation of aeronautic or space systems; it is therefore used by several Onera departments. By offering its expertise in modeling, methods development and implementation, and an expertise covering many different branches of physics, the DTIM is fully playing its part in this area of Onera's work.

On this theme, the department is interested in new computing techniques, as much in terms of hardware as software, and in the use or reuse of new tools (adaptive grids, code coupling).

This activity is leading to the production of robust, portable and long-life software, a necessity in the face of the rapid development of hardware (even sometimes its disappearance).

New challenges

This discipline is having to confront new requirements as much in terms of the richness and complexity of the phenomena to be simulated, having then to take into account multi-scale or multi-physical aspects, as in terms of the exploiting the ever increasing parallelism of scientific computers.

The existence of codes that simulate physical phenomena precisely and rapidly opens up new possibilities for research in the field of optimization and control.

Thematic orientations

Research subjects

With the development of basic techniques, spanning the various objectives of the scientific branches of Onera, the DTIM can offer expertise at all levels, from mathematical modeling to numerical simulation:

Mathematical modeling is an activity that the department is involved in more and more. This is particularly the case in:

the coupling of multi-physical models such as the interaction of a flow with a plasma or fluid-structure coupling
the identification and control of complex physical phenomena with partial differential equations

Numerical algorithms for the study of high precision methods for systems of wave propagation equations, the creation of parallel solvers adapted to large linear systems, code coupling, multi-scale numerical modeling and the building of software packages for the adaptation of non-structured grids.

Optimization methodologies for the multi-disciplinary optimization of physical systems modeled with calculation codes.
Parallel and distributed processing for the development of numerical methods for massively parallel architectures and the production of software tools for code coupling.