Doklady Belorusskogo gosudarstvennogo universiteta informatiki i radioèlektroniki (Jun 2021)
Verification of numerical methods and mathematical model developed for simulation of radionuclides migration in natural disperse environments
Abstract
Simulation of radionuclides vertical migration was successfullyimplemented in the first versions of SPS (Simulation of Processes in Soil) software and was based on a numerical solution of the mathematical model of interconnected heat and moisture transfer in one dimension. But in order to solve problems of a comprehensive assessment of the state of the biosphere under radionuclides pollution and better approximation of simulation results to real processes, authors developed SPS v2.0 software. One of the modules of SPS v2.0 uses new mathematical model that describes the spatial migration of radionuclides in soil (3D-model). The numerical solution of this mathematical model is based on the application of the finite element method and the analytical approximation of thermal conductivity and liquid pressure coefficients. Such approach makes possible to use parallel computing technologies for simulation. The mathematical model used in SPS v2.0, as well as numerical methods forsolving it, require verification, which is carried out in this article. Verification of the developed numerical methodswas carried out using ComsolMultiphysics software and SPS v2.0 module with the following comparison of the calculation results. The difference in the calculation results obtained using the listed software is less than 5 %, therefore, the numerical methods are correctly implemented in SPS v2.0 and have a solution accuracy comparable to the numerical methods used in modern software. For the mathematical model verification were used the results of experimental measurements of meteorological conditions, distribution of moisture and temperature in soil, which were compared with the simulation results obtained in SPS v2.0. The comparison shows that the error in the calculation of the analyzed parameters does not exceed 5 %, which allows the developed model to be used for solvation of practical problems in the subject area.
