Parameter-uniform numerical methods for some singularly perturbed nonlinear initial value problems

Parameter-uniform numerical methods for singularly perturbed nonlinear scalar initial value problems are both constructed and analysed in this paper. The conditions on the initial condition for a stable initial layer to form are identified. The character of a stable initial layer in the vicinity of a double root of the reduced algebraic problem is different to the standard layer structures appearing in the neighbourhood of a single stable root of the reduced problem. Results for a problem where two reduced solutions intersect are also discussed. Numerical results are presented to illustrate the theoretical results obtained.     

Hybrid numerical methods for periodic initial value problems involving second-order differential equations

Computer simulation of problems in celestial mechanics often leads to the numerical solution of the system of second-order initial value problems with periodic solutions. When conventional methods are applied to obtain the solution, the time increment must be limited to a value of the order of the reciprocal of the frequency of the periodic solution.In this paper hybrid methods of orders four and six which are P-stable are developed. Further, the adaptive hybrid methods of polynomial order four and trigonometric order one have also been discussed. The numerical results for the undamped Duffing equation with a forced harmonic function are listed.     

E-stable methods for exponentially decreasing solutions of second order initial value problems

In this paperE-stable methods ofO(h ⁴),O(h ⁸) andO(h ¹²) are derived for the direct numerical integration of initial value problems of second order differential equations with exponentially decreasing solutions. Numerical results are presented for both linear and nonlinear problems.     

High-order one-step P-stable methods for the numerical integration of periodic initial value problems

Using Lobatto nodes, one-step methods of order six and eight have been obtained for the second-order differential equation y″ = f(x, y), y(x₀) = y₀, y′(x₀) = y′₀. The methods are shown to be P-stable. If

, then at each integration step a system of dimension 3s, 4s, respectively, has to be solved. The numerical results, for two problems, obtained by using these methods are given in the end.     

A multi-point numerical integrator with trigonometric coefficients for initial value problems with periodic solutions

Based on the collocation technique, we introduced a unifying approach for deriving a family of multi-point numerical integrators with trigonometric coefficients for the numerical solution of periodic initial value problems. A practical 3-point numerical integrator was presented, whose coefficients are generalizations of classical linear multistep methods such that the coefficients are functions of an estimate of the angular frequency ω. The collocation technique yields a continuous method, from which the main and complementary methods are recovered and expressed as a block matrix finite difference formula that integrates a second-order differential equation over non-overlapping intervals without predictors. Some properties of the numerical integrator were investigated and presented. Numerical examples are given to illustrate the accuracy of the method.     

EfficientP-stable methods for periodic initial value problems

Efficient families ofP-stable formulae are developed for the numerical integration of periodic initial value problems where the required solution has an unknown period. Formulae of orders 4 and 6 requiring respectively 2 and 4 function evaluations per step are derived and some numerical results are given.     

A numerical comparison of Chebyshev methods for solving fourth order semilinear initial boundary value problems

In solving semilinear initial boundary value problems with prescribed non-periodic boundary conditions using implicit-explicit and implicit time stepping schemes, both the function and derivatives of the function may need to be computed accurately at each time step. To determine the best Chebyshev collocation method to do this, the accuracy of the real space Chebyshev differentiation, spectral space preconditioned Chebyshev tau, real space Chebyshev integration and spectral space Chebyshev integration methods are compared in the L² and W^2,2 norms when solving linear fourth order boundary value problems; and in the L^∞([0,T];L²) and L^∞([0,T];W^2,2) norms when solving initial boundary value problems. We find that the best Chebyshev method to use for high resolution computations of solutions to initial boundary value problems is the spectral space Chebyshev integration method which uses sparse matrix operations and has a computational cost comparable to Fourier spectral discretization.     

Blended extended linear multistep methods for the accurate numerical integration of stiff initial value problems

A class of blended extended linear multistep methods suitable for the approximate numerical integration of stiff systems of first order ordinary differential equations is described. These methods are formulated as a result of combining the second derivative extended backward differentiation formulae of Cash and the blended linear multistep methods of Skeel and Kong. The new methods combine a high order or accuracy with good stability properties and, as a direct consequence, they are often suitable for the numerical integration of stiff differential systems when high accuracy is requested. In the first part of the present paper we consider the derivation of these new blended methods and give the coefficients and stability regions for formulae of order up to and including 10. In the second half we consider their practical implementation. In particular we describe a variable order/variable step package based on these blended formulae and we evaluate the performance of this package on the well known DETEST test set. It is shown that the new code is reliable on this test set and is competitive with the well known second derivative method of Enright.     

On the numerical integration of nonlinear initial value problems by linear multistep methods

We study the numerical solution of the nonlinear initial value problem $$\left\{ {\begin{array}{*{20}c} {{{du(t)} \mathord{\left/ {\vphantom {{du(t)} {dt}}} \right. \kern-\nulldelimiterspace} {dt}} + Au(t) = f(t),t > 0} \\ {u(0) = c,} \\ \end{array} } \right.$$ whereA is a nonlinear operator in a real Hilbert space. The problem is discretized using linear multistep methods, and we assume that their stability regions have nonempty interiors. We give sharp bounds for the global error by relating the stability region of the method to the monotonicity properties ofA. In particular we study the case whereAu is the gradient of a convex functional φ(u).     

Piecewise polynomial Taylor methods for initial value problems

A class of explicit Taylor-type methods for numerically solving first-order ordinary differential equations is presented. The basic idea is that of generating a piecewise polynomial approximating function, with a given order of differentiability, by repeated Taylor expansion. Sharp error bounds for the approximation and its derivatives are given along with a stability analysis.This work was supported by the United States Atomic Energy Commission.     

Special boundedness properties in numerical initial value problems

For Runge-Kutta methods, linear multistep methods and other classes of general linear methods much attention has been paid in the literature to important nonlinear stability properties known as total-variation-diminishing (TVD), strong stability preserving (SSP) and monotonicity. Stepsize conditions guaranteeing these properties were studied by Shu and Osher (J. Comput. Phys. 77:439–471, 1988) and in numerous subsequent papers. Unfortunately, for many useful methods it has turned out that these properties do not hold. For this reason attention has been paid in the recent literature to the related and more general properties called total-variation-bounded (TVB) and boundedness.     

Splitting methods for second‐order initial value problems

We consider implicit integration methods for the solution of stiff initial value problems for second-order differential equations of the special form y' = f(y). In implicit methods, we are faced with the problem of solving systems of implicit relations. This paper focuses on the construction and analysis of iterative solution methods which are effective in cases where the Jacobian of the right‐hand side of the differential equation can be split into a sum of matrices with a simple structure. These iterative methods consist of the modified Newton method and an iterative linear solver to deal with the linear Newton systems. The linear solver is based on the approximate factorization of the system matrix associated with the linear Newton systems. A number of convergence results are derived for the linear solver in the case where the Jacobian matrix can be split into commuting matrices. Such problems often arise in the spatial discretization of time‐dependent partial differential equations. Furthermore, the stability matrix and the order of accuracy of the integration process are derived in the case of a finite number of iterations. This revised version was published online in August 2006 with corrections to the Cover Date.     

Contractivity in the numerical solution of initial value problems

Numerische Mathematik - Consider a linear autonomous system of ordinary differential equations with the property that the norm |U(t)| of each solutionU(t) satisfies |U(t)|≦|U(0)| (t≧0)....     

Continuous numerical solution of coupled parabolic initial value problems

This paper deals with the construction of continuous numerical solutions of coupled parabolic initial value problems using Fer's factorization and the Fourier transform approach.     

Inverse linear multistep methods for the numerical solution of initial value problems of second order differential equations

In this paper inverse linear multistep methods for the numerical solution of second order differential equations are presented. Local accuracy and stability of the methods are defined and discussed. The methods are applicable to a class of special second order initial value problems, not explicitly involving the first derivative. The methods are not convergent, but yield good numerical results if applied to problems they are designed for. Numerical results are presented for both the linear and nonlinear initial value problems.     

Recent advances in methods for numerical solution of O.D.E. initial value problems

This is a review paper which describes recent advances in numerical methods and computer codes for solving initial value problems of ordinary differential equations. Particular emphasis is placed upon stiff systems.     

Convergence on error correction methods for solving initial value problems

Higher-order semi-explicit one-step error correction methods(ECM) for solving initial value problems are developed. ECM provides the excellent convergence O(h^2p+2)$O\left(h^{2p+2}\right)$ one wants to get without any iteration processes required by most implicit type methods. This is possible if one constructs a local approximation having a residual error O(h^p)$O\left(h^p\right)$ on each time step. As a practical example, we construct a local quadratic approximation. Further, it is shown that special choices of parameters for the local quadratic polynomial lead to the known explicit second-order methods which can be improved into a semi-explicit type ECM of the order of accuracy 6$6$. The stability function is also derived and numerical evidences are presented to support theoretical results with several stiff and non-stiff problems. It should be remarked that the ECM approach developed here does not yield explicit methods, but semi-implicit methods of the Rosenbrock type. Both ECM and Rosenbrock’s methods require to solve a few linear systems at each integration step, but the ECM approach involves 2p+2$2p+2$ evaluations of the Jacobian matrix per integration step whereas the Rosenbrock method demands one evaluation only. However, it is much easier to get high order methods by using the ECM approach.     

Monotonicity and finite element methods for hyperbolic initial value problems

A well known theórem about super- and subfunctions for the solution of hyperbolic initial value problems constructs differentiable functions as upper and lower bounds (see Walter [1], 21 XIII). The proof can be done by transforming the differential equation problem into a set of integral equations, using the monotonicity-properties of the arising integral operators. This proof needs an integral representation for twice differentiable functions. It is shown that this proceeding can be generalized to get upper and lower bounds in terms of finite element functions. To do this, we give an integral representation for continuous, piecewise differentiable functions, including the discontinuities of their derivatives. Then the generalization of the classical proof yields interface conditions for the finite element functions. Finally, it is demonstrated how to realize numerically these conditions.