Download e-book for iPad: Applied Iterative Methods by Louis A. Hageman

By Louis A. Hageman

ISBN-10: 0123133408

ISBN-13: 9780123133403

This graduate-level textual content examines the sensible use of iterative equipment in fixing huge, sparse structures of linear algebraic equations and in resolving multidimensional boundary-value difficulties. Assuming minimum mathematical historical past, it profiles the relative benefits of a number of common iterative strategies. themes comprise polynomial acceleration of uncomplicated iterative equipment, Chebyshev and conjugate gradient acceleration methods appropriate to partitioning the linear method right into a “red/black” block shape, adaptive computational algorithms for the successive overrelaxation (SOR) technique, and computational facets within the use of iterative algorithms for fixing multidimensional difficulties. 1981 ed. forty eight figures. 35 tables.

Show description

Read or Download Applied Iterative Methods PDF

Best mathematical analysis books

Read e-book online The Immersed Interface Method: Numerical Solutions of PDEs PDF

Interface difficulties come up whilst there are diversified fabrics, comparable to water and oil, or an identical fabric at varied states, equivalent to water and ice. If partial or traditional differential equations are used to version those functions, the parameters within the governing equations are usually discontinuous around the interface setting apart the 2 fabrics or states, and the resource phrases are frequently singular to reflect source/sink distributions alongside codimensional interfaces.

New PDF release: Statistics of Random Processes: I. General Theory

The topic of those volumes is non-linear filtering (prediction and smoothing) thought and its software to the matter of optimum estimation, keep an eye on with incomplete info, info idea, and sequential checking out of speculation. the necessary mathematical heritage is gifted within the first quantity: the idea of martingales, stochastic differential equations, absolutely the continuity of likelihood measures for diffusion and Ito procedures, components of stochastic calculus for counting tactics.

An Introduction to Fourier Analysis and Generalised - download pdf or read online

This monograph on generalised services, Fourier integrals and Fourier sequence is meant for readers who, whereas accepting thought the place every one element is proved is healthier than one in line with conjecture, however search a remedy as effortless and loose from issues as attainable. Little precise wisdom of specific mathematical recommendations is needed; the booklet is acceptable for complex collage scholars, and will be used because the foundation of a brief undergraduate lecture path.

Additional info for Applied Iterative Methods

Example text

13) in the more computationally convenient form ,,_,. Λ - ( I - * · ) - . 215) n>2, where σ = l/w(l) = (M(G) - m(G))/(2 - M(G) - m(G)). 2 49 OPTIMAL CHEBYSHEV ACCELERATION It can also be shown (Varga, 1962) that limp„ EE pa = 2/(1 + V l - σ2). 17) We now examine the convergence rate of the optimal Chebyshev procedure. )/(l + V1 - ^2)· (4-219) Thus we obtain S(P„(G)) = 2r" /2 /(l + r"). 17) can be expressed in the form Rao(Pn(G))= - i l o g r . 19) that r = p^ — 1. Thus we also have that lim iS(Pn(G))Vln = (pn - 1) 1/2 and that R^P^G)) = - \ logip, - 1).

3) p B + ! 11) with y and pn+i replacing the y and p n + 1 , respectively. 1, the iterates for the polynomial acceleration procedure based on P„jE(x) ma Y be given by uin+1) = Pn+1{y(Gu™ + fc) + (1 - y)u™} + (1 - pn+Mn~l). , and " - ' · " - < ' - « > - . 8) In the above discussion, we have tacitly assumed that ME Φ mE. If ME = mE,theny = 1/(1 — ME\px = p2 = · · · = landσ E = 0. 13). We shall not consider this case separately in the balance of this section. The correct formulas can be obtained by a limiting process.

Let such a matrix Q be defined by the splitting A = Q - R for the coefficient matrix A. 1) X l with the corresponding iteration matrix G = I — Q~ A = Q~ R. We now consider those methods for which the matrix Q has the form Q = HK. " Moreover, to maximize the rate of convergence, H and K should also be chosen such that the spectral radius of the iteration matrix G is as small as possible. If A is symmetric and positive definite, one choice is to let H = LT and K = L, where L is the upper triangular matrix defined by the Cholesky decomposition A = LTL of A.

Download PDF sample

Applied Iterative Methods by Louis A. Hageman


by John
4.4

Rated 4.77 of 5 – based on 32 votes

About admin