Romain Hild

Research Engineer

• MOR_DICUS: Conception of an open-source library for industrial use of model reduction methods. Project in collaboration with EDF R\&D, SAFRAN, École des Mines de Paris, ... Application of reduced basis methods in Python and C++
• Ibat: Use of Building Information Model (BIM) to develop efficient heat exchange models combining physical models, IA, and data coming from sensors deplyed in the building. Creation of micro services deploying a REST API with Node.js, use of missing data imputation methods, supervision of internships
• Eye2Brain: Development of mathematical methods to describe relations between eye and brain for diagnosis purposes. 3D-0D coupling between Feel++ and OpenModelica with HDG methods

Research Engineer

• Non standard resolution method for Navier-Stokes equations with specific boundary conditions, using a spectral basis.
• Parallel computing and HPC
• Linear algebra and algorithmic

Internship

Parallelisation of a code for aerodynamical and instationary computation and validation by comparison with existing data.

Internship

A contrario approach for image processing to object indexation

Université de Strasbourg

Optimization and control of high field magnets

• Resolution of non linear coupled problems in industrial context, involving thermo-electric, magnetostatic and elasticity problems.
• Use of Contiunous Galerkin and Hybridizable Discontinuous Galerkin methods
• Use of Reduced Basis method for model order reduction and use of (Discrete) Empirical Interpolation Method (DEIM and EIM) with Simultaneous EIM and RB (SER) method to deal with non linearity.
• Geometrical optimization of a magnet to obtain a better homogeneity of the magnetic field, using RB method and Empirical Quadrature Method (EQM)