RUS  ENG
Full version
JOURNALS // Vestnik Yuzhno-Ural'skogo Universiteta. Seriya Matematicheskoe Modelirovanie i Programmirovanie // Archive

Vestnik YuUrGU. Ser. Mat. Model. Progr., 2013 Volume 6, Issue 4, Pages 5–14 (Mi vyuru99)

This article is cited in 1 paper

Mathematical Modelling

On Stable Algorithms for Numerical Solution of Integral-Algebraic Equations

M. V. Bulatovab, O. S. Budnikovac

a Institute for System Dynamics and Control Theory of Siberian Branch of Russian Academy of Sciences
b Irkutsk State Technical University, Irkutsk, Russian Federation
c East Siberian State Academy of Education, Irkutsk

Abstract: There is the necessity to study integral-algebraic equations if a prototype process has an aftereffect at the analysis of various areas of science. Particularly, a system of interrelated Volterra equations of the first and second kind and algebraic equations can be written as integral-algebraic equation. In this paper linear integral-algebraic equations are considered. We have constructed multistep methods for numerical solutions of IAEs. These methods are based on Adams quadrature formulas and on extrapolation formulas as well. We have proven suggested algorithms convergence. In this paper we show that our multistep methods have a property of self-regularizing; and regularization parameter is the step of a grid, which is connected with the level of accuracy of right-part error of the system under consideration. The results of numerical experiments illustrate theoretical computations.

Keywords: integral-algebraic equations; multistep methods; self-regularization.

UDC: 519.62

MSC: 65R20

Received: 02.07.2013



© Steklov Math. Inst. of RAS, 2024