I'm Alexis Gobé. I am currently a PhD student in the region of Nice, France.
I started my PhD at the Inria Sophia Antipolis research center in November 2016 in the Nachos project-team under the supervision of Stéphane Lanteri, head of Nachos project team.
My PhD thesis is about numerical methods for nanophotonics, specificaly for photovoltaic problems.
The two aimed methods are the Discontinuous Galerkin Time Domain (DGTD) method and a new family of high-order multiscale finite element method called Multiscale Hybrid-Mixed (MHM) method.
Find below more information about myself as well as a sample of my past work. To contact me, find links at the bottom of the page.
I worked on different strategies to reduce the load balancing when dealing with highly complicated configuration (local refinement, hybrid meshes, local observables, ...).
Development of a MHM Mesh generator for complex configuration by exploiting CAD features from GMSH 3.0+. This is the first step for the implementation of the 3D MHM-DGTD solver.
We study light trapping in a silicon-based thin-film solar cell setup that consists of several randomly textured layers using DGTD solver DIOGENeS5+. The focus is on a-Si:H5 and µc-Si:H5* which belong to the family of disordered semiconductors. More information about this case on Nachos highlights of the month.
5. Amorphous silicon
5*. Microcrystalline silicon
5+. DIscOntinuous GalErkin Nanoscale Solvers
Work on the MHM-DGTD4 code of Raphaël Léger for the 2D time domain Maxwell's equations. Developpement of a tool for the generation of the coarse and of the local meshes of complex configuration. Implementation of local order per edges to deal with heterogeneous media.
Here we are interested by solar absorption in ultra-thin-film GaAs3 photonic crystals. GaAs solar cells offer an alternative material to mostly used silicon-based solar cells.
3. Gallium arsenide
Worked on the HORSE1 platform of Ludovic Moya , platform based on HDG1* method for solving the time-harmonic Maxwell’s equation in 3D and add into it the possibility to use tetrahedral/hexahedral hybrid meshes.
1. High Order solver for Radar cross Section Evaluation
1*. Hybrid Discontinuous Galerkin Methods