Parallel efficient ModulAr tools for multiPhysics and multIscale complEx simulations. Shock waves propagation in complex media – MAPIE

Simulation and modeling of shock waves propagation in complex media is an challenging issue, belonging to complex multiphysics problems, that has a wide range of applications ranging from astrophysics, gas dynamics, structural dynamics, materials sciences to the next generation of jet propulsion. The purpose of MAPIE project is to develop improved computational methods for solving multiscale, multiphysics problems using high-performance computing. The focus is on shock waves propagation in complex media including, in fluid mechanics domain, compressible turbulence and detonations, in linear structure dynamics domain, HF (high frequency) wave propagation and attenuation in heterogeneous materials (polycrystalline material) or cellular solids (honeycomb or foams), or in more severe cases, high-speed multi-material impact and penetration resulting in highly non linear structure deformation. The modular concept will be adopted based on solution-adaptive codes that treat physical models with numerics suited to each phenomenon. The development and applications of different models and methods will be based on the high-order structured finite differences large-eddy simulation hydrocode CHOC-WAVES, developed at CORIA CNRS and the object-oriented finite elements structure code OOFE developed at MSSMat CNRS.

The computational requirements for solving both fluid and structure problems will be achieved by exploiting the full potential of available French massively parallel computers (within GENCI projects). The objective is to highlight the programming efficiency, the portability and the performance issues of the existing and/or the new algorithms in high-performance computing large-scale environments.

The efficient physical models implementation as well as the different numerical experiments that will be conduced in the framework of this project along with the data post-processing and the flow visualization will lead to dramatic improvements in the fidelity of the results and allow to better gain in efficiency concerning flexibility and, therefore, modeling time and cost.

One of the main goal of the current proposal is, on the one hand, to improve the existing numerical tools and to develop new methods for numerical modeling and computational physics of wave propagation phenomena in complex media for respectively fluid and structure applications and, on the other hand, to develop efficient algorithms to couple fluid and structure simulations.