Future high-efficiency plants converting concentrated solar energy into electricity or solar fuels will require a heat receiver in which thermal transfers will be optimized to achieve maximum conversion efficiency and high temperatures (> 700 ° C). Among the technologies studied, porous ceramic receivers and exchangers have the advantage of resisting high temperatures and of increasing transfers between the heat transfer fluid and the solid matrix. In order to achieve high thermal efficiency and to improve the understanding of limiting phenomena, the optimization and intensification of heat transfer requires an accurate model of the conversion efficiency dependencies to the geometric, thermophysical and thermoradiative properties of the porous solid phase. The thesis aims to develop a reduced model of the porous medium whose parameters will be obtained through the analysis of the random paths used by the Monte Carlo Symbolic (MCS) statistical method to solve coupled thermal transfers. The model will then be used to optimize the receivers and the porous exchangers according to the operating conditions and their geometries.