Displaying 101 – 120 of 1948

Showing per page

A conjugate gradient method with quasi-Newton approximation

Jonas Koko (2000)

Applicationes Mathematicae

The conjugate gradient method of Liu and Storey is an efficient minimization algorithm which uses second derivatives information, without saving matrices, by finite difference approximation. It is shown that the finite difference scheme can be removed by using a quasi-Newton approximation for computing a search direction, without loss of convergence. A conjugate gradient method based on BFGS approximation is proposed and compared with existing methods of the same class.

A conservative spectral element method for the approximation of compressible fluid flow

Kelly Black (1999)

Kybernetika

A method to approximate the Euler equations is presented. The method is a multi-domain approximation, and a variational form of the Euler equations is found by making use of the divergence theorem. The method is similar to that of the Discontinuous-Galerkin method of Cockburn and Shu, but the implementation is constructed through a spectral, multi-domain approach. The method is introduced and is shown to be a conservative scheme. A numerical example is given for the expanding flow around a point...

A continuous finite element method with face penalty to approximate Friedrichs' systems

Erik Burman, Alexandre Ern (2007)

ESAIM: Mathematical Modelling and Numerical Analysis

A continuous finite element method to approximate Friedrichs' systems is proposed and analyzed. Stability is achieved by penalizing the jumps across mesh interfaces of the normal derivative of some components of the discrete solution. The convergence analysis leads to optimal convergence rates in the graph norm and suboptimal of order ½ convergence rates in the L2-norm. A variant of the method specialized to Friedrichs' systems associated with elliptic PDE's in mixed form and reducing the number...

A contribution to Runge-Kutta formulas of the 7th order with rational coefficients for systems of differential equations of the first order

Anton Huťa, Vladimír Penjak (1984)

Aplikace matematiky

The purpose of this article is to find the 7th order formulas with rational parameters. The formulas are of the 11th stage. If we compare the coefficients of the development i = 1 h i i ! d i - 1 d x i - 1 𝐟 x , 𝐲 ( x ) up to h 7 with the development given by successive insertion into the formula h . f i ( k 0 , k 1 , ... , k i - 1 ) for i = 1 , 2 , ... , 10 and k = i = 0 10 p i , k i we obtain a system of 59 condition equations with 65 unknowns (except, the 1st one, all equations are nonlinear). As the solution of this system we get the parameters of the 7th order Runge-Kutta formulas as rational numbers.

Currently displaying 101 – 120 of 1948