On modeling sources of radiation-induced effects in heterogeneous materials

V. A. Soboleva, M. E. Zhukovskiy

Keldysh Institute of Applied Mathematics of RAS

Abstract: An approach to the calculation of initial data for computing the radiation-induced secondary effects in a heterogeneous environment is considered. A method for solving the problem of integration "according to data" of the results of modeling cascade processes of radiation transport and the processes of generation of secondary radiation-induced effects is proposed. The method is based on a multidimensional approximation of the results of statistical modeling of the interaction of radiation with matter on a difference grid designed for numerical solution of the equations of electro- and thermodynamics. The approximation is built using neural network technology. The geometric model of a heterogeneous medium is based on Stilinger-Lubachevsky algorithms for multimodal structures. The results of demonstration calculations are presented.

Keywords: radiation transfer, multidimensional approximation, radiation-induced effects, neural networks.

Received: 16.11.2021
Revised: 16.11.2021
Accepted: 17.01.2022

DOI: 10.20948/mm-2022-03-07

 English version: Mathematical Models and Computer Simulations, 2022, 14:5, 829–836