


Vol 54, No 7 (2018)
- Year: 2018
- Articles: 15
- URL: https://journals.rcsi.science/0012-2661/issue/view/9350
Numerical Methods
Unconditionally Stable Algorithm for Solving the Three-Dimensional Nonstationary Navier–Stokes Equations
Abstract
We propose an unconditionally stable method for solving the three-dimensional nonstationary Navier–Stokes equations in the velocity–pressure variables. The method is based on a conservative finite-difference scheme and the simultaneous solution of the momentum and continuity equations at each time layer. The velocity and pressure fields are calculated by using a parallel algorithm for solving systems of linear equations by the Gauss method.



Finite Element Models of Hyperelastic Materials Based on a New Strain Measure
Abstract
To construct constitutive equations for hyperelastic materials, one increasingly often proposes new strain measures, which result in significant simplifications and error reduction in experimental data processing. One such strain measure is based on the upper triangular (QR) decomposition of the deformation gradient. We describe a finite element method for solving nonlinear elasticity problems in the framework of finite strains for the case in which the constitutive equations are written with the use of the QR-decomposition of the deformation gradient. The method permits developing an efficient, easy-to-implement tool for modeling the stress–strain state of any hyperelastic material.



Convergence of a Projection-Difference Method for the Approximate Solution of a Parabolic Equation with a Weighted Integral Condition on the Solution
Abstract
In a Hilbert space, we consider an abstract linear parabolic equation defined on an interval with a nonlocal weighted integral condition imposed on the solution. This problem is solved approximately by a projection-difference method with the use of the implicit Euler method in the time variable. The approximation to the problem in the spatial variables is developed with the finite element method in mind. An estimate of the approximate solution is obtained, the convergence of the approximate solutions to the exact solution is proved, and the error estimates, as well as the orders of the rate of convergence, are established.



Adaptive Interpolation Algorithm Based on a kd-Tree for Numerical Integration of Systems of Ordinary Differential Equations with Interval Initial Conditions
Abstract
We consider issues related to the numerical solution of interval systems of ordinary differential equations. We suggest an algorithm that permits finding interval estimates of solutions with prescribed accuracy in reasonable time. The algorithm constructs an adaptive partition (a dynamic structured grid) based on a kd-tree over the space formed by interval initial conditions for the ordinary differential equations. In the operation of the algorithm, a piecewise polynomial function interpolating the dependence of the solution on the specific values of interval parameters is constructed at each step of solution of the original problem. We prove that the global error estimate linearly depends on the height of the kd-tree. The algorithm is tested on several examples; the test results show its efficiency when solving problems of the class under study.



Quasi-One-Dimensional Flow of a Fluid with Anisotropic Viscosity in a Pulsating Vessel
Abstract
The problem on a fluid flow in a pulsating vessel is considered in the framework of the quasi-one-dimensional hemodynamic equations. The fluid viscosity is assumed to be anisotropic; i.e., the viscosity coefficient depends on the flow direction. The possible solutions are studied analytically and numerically.



Coefficient Stability of the Solution of a Difference Scheme Approximating a Mixed Problem for a Semilinear Parabolic Equation
Abstract
We study the coefficient stability of a difference scheme approximating a mixed problem for a one-dimensional semilinear parabolic equation. We obtain sufficient conditions on the input data under which the solutions of the differential and difference problems are bounded. We also obtain estimates of perturbations of the solution of a linearized difference scheme with respect to perturbations of the coefficients; these estimates agree with the estimates for the differential problem.



Construction and Structure Properties of Solutions of a Periodic Boundary Value Problem for a Generalization of the Matrix Lyapunov and Riccati Equations
Abstract
We obtain constructive sufficient conditions for the unique solvability of a periodic boundary value problem for a matrix differential equation that generalizes the Lyapunov and Riccati equations, develop an algorithm for constructing the solution of this equation, estimate the domain where the solution is localized, and study the structural properties of the solution.



Parametrization of the Solution of the Kepler Problem and New Adaptive Numerical Methods Based on This Parametrization
Abstract
We propose a one-parameter family of adaptive numerical methods for solving the Kepler problem. The methods preserve the global properties of the exact solution of the problem and approximate the time dependence of the phase variables with the second or fourth approximation order. The variable time increment is determined automatically from the properties of the solution.



Numerical Solution of Inverse Problems for a Hyperbolic Equation with a Small Parameter Multiplying the Highest Derivative
Abstract
We consider two inverse problems for a hyperbolic equation with a small parameter multiplying the highest derivative. The inverse problems are reduced to systems of linear Volterra integral equations of the second kind for the unknown functions. These systems are used to prove the existence and uniqueness of the solution of the inverse problems and numerically solve them. The applicability of the methods developed here to the approximate solution of the problem on an unknown source in the heat equation is studied numerically.



Development and Application of an Exponential Method for Integrating Stiff Systems Based on the Classical Runge–Kutta Method
Abstract
We study numerical methods for solving stiff systems of ordinary differential equations. We propose an exponential computational algorithm which is constructed by using an exponential change of variables based on the classical Runge–Kutta method of the fourth order. Nonlinear problems are used to prove and demonstrate the fourth order of convergence of the new method.



Generalized Collocation Method for Integro-Differential Equations in an Exceptional Case
Abstract
We study a linear integro-differential equation with a coefficient that has finite-order zeros. To solve the equation approximately in a distribution space, we suggest and substantiate a generalized collocation method based on special interpolation polynomials.



Locally One-Dimensional Difference Scheme for the Third Boundary Value Problem for a Parabolic Equation of the General Form with a Nonlocal Source
Abstract
We consider a locally one-dimensional scheme for an equation of parabolic type of the general form in a p-dimensional parallelepiped, obtain an a priori estimate for its solution, and prove that the solutions of this scheme converge to a solution of the equation at the rate O(|h|2 + τ), where |h|2 = h12 + · · · + hp2 and pα, α = 1,..., p, and τ are the steps in the space and time variables. We do not assume that the operator in the leading part of the equation is sign definite.



Ordinary Differential Equations
Stability of Differential-Algebraic Equations under Uncertainty
Abstract
We consider a linear time-invariant homogeneous system of first-order ordinary differential equations with a noninvertible matrix multiplying the derivative of the unknown vector function and with perturbed coefficients. We introduce a class of perturbations of the coefficient matrices of the system and determine conditions on the perturbations of this class under which they do not affect the internal structure of the system. We obtain sufficient conditions for the robust stability of the system under such perturbations.






Some Properties of the Angles between Lineals of Solutions of Exponentially Dichotomous Systems
Abstract
We prove that the maximum angle between two arbitrary unstable solution subspaces of an exponentially dichotomous finite-dimensional system on the half-line tends to zero as the time variable tends to infinity. In particular, there exists a positive constant such that the angle between an arbitrary unstable solution subspace and the stable lineal of the system is greater than this constant for all sufficiently large times.


