On the solution of the neutron diffusion kinetic equation in planar geometry free of stiffness with convergence analysis
Autor: | Marco T. Vilhena, Fernanda Tumelero, Bardo E. J. Bodmann, Celso Marcelo Franklin Lapa |
---|---|
Rok vydání: | 2019 |
Předmět: |
Recursion
Diffusion equation 020209 energy Mathematical analysis 02 engineering and technology 01 natural sciences 010305 fluids & plasmas symbols.namesake Nuclear Energy and Engineering Homogeneous differential equation Ordinary differential equation 0103 physical sciences Diagonal matrix 0202 electrical engineering electronic engineering information engineering Taylor series symbols Boundary value problem Coefficient matrix Mathematics |
Zdroj: | Annals of Nuclear Energy. 125:272-282 |
ISSN: | 0306-4549 |
DOI: | 10.1016/j.anucene.2018.11.024 |
Popis: | In this work we solve the space kinetic diffusion equation in a one-dimensional geometry considering a homogeneous domain, for two energy groups and six groups of delayed neutron precursors. The proposed methodology makes use of a Taylor expansion in the space variable of the scalar neutron flux (fast and thermal) and the concentration of delayed neutron precursors, allocating the time dependence to the coefficients. Upon truncating the Taylor series at quadratic order, one obtains a set of recursive systems of ordinary differential equations, where a modified decomposition method is applied. The coefficient matrix is split into two, one constant diagonal matrix and the second one with the remaining time dependent and off-diagonal terms. Moreover, the equation system is reorganized such that the terms containing the latter matrix are treated as source terms. Note, that the homogeneous equation system has a well known solution, since the matrix is diagonal and constant. This solution plays the role of the recursion initialization of the decomposition method. The recursion scheme is set up in a fashion where the solutions of the previous recursion steps determine the source terms of the subsequent steps. A second feature of the method is the choice of the initial and boundary conditions, which are satisfied by the recursion initialization, while from the first recursion step onward the initial and boundary conditions are homogeneous. The recursion depth is then governed by a prescribed accuracy for the solution. To this end we present error estimates and a convergence and stability analysis. |
Databáze: | OpenAIRE |
Externí odkaz: |