Thanks for your reply.
I combined radiation + convection as you said.
But my results still show me the ambiant temperature on the surfaces, which is not accurate (especially with 973°K within the plate, it should be more like 500°K I guess).
Do I have to modelize the ambiant air around my...