# Numerical Hamiltonian Problems (Applied Mathematics and by J. M. Sanz-Serna, M. P. Calvo

By J. M. Sanz-Serna, M. P. Calvo

Similar computational mathematicsematics books

Numerical Methods for Structured Matrices and Applications: The Georg Heinig Memorial Volume

This cross-disciplinary quantity brings jointly theoretical mathematicians, engineers and numerical analysts and publishes surveys and study articles regarding the subjects the place Georg Heinig had made awesome achievements. specifically, this comprises contributions from the fields of established matrices, quickly algorithms, operator concept, and purposes to method conception and sign processing.

Robust Algebraic Multilevel Methods and Algorithms (Radon Series on Computational and Applied Mathematics)

This booklet bargains with algorithms for the answer of linear structures of algebraic equations with large-scale sparse matrices, with a spotlight on difficulties which are got after discretization of partial differential equations utilizing finite point tools. offers a scientific presentation of the hot advances in strong algebraic multilevel tools.

Extra resources for Numerical Hamiltonian Problems (Applied Mathematics and Mathematical Computation, No 7)

Sample text

Let h = 1/ N be the mesh size, and let xi = ih (0 i N ) be the mesh points. For convenience, we introduce the ﬁnite difference operators δh2 and Ph as follows: δh2 u (xi ) = u (xi −1 ) − 2u (xi ) + u (xi +1 ), h Ph u (xi ) = 2 12 1 N − 1, i u (xi −1 ) + 10u (xi ) + u (xi +1 ) , 1 i N − 1. 7) Using the following Numerov’s formula (cf. 6) that −δh2 v (xi ) + Ph γ (xi ) v (xi ) = Ph g xi , u (xi ) + O h6 , v (x0 ) = T (α ), T u (xi ) = v (xi ), v (x N ) = T (β). 10) where u i and v i represent the approximations of u (xi ) and v (xi ), respectively, and γi = γ (xi ), g i (u i ) = g (xi , u i ).

1) and −δh2 v i = Ph f (xi , u i ), v 0 = T (α ), T (u i ) = v i , 1 i N − 1, v N = T (β). 2), we show the convergence of (u i , v i ) to (u (xi ), v (xi )) as h → 0. -M. 1. Assume that u i , u (xi ) ∈ Λ ⊂ I for each i and some interval Λ in R, and assume f u (x, ξ ) k(η) < π 2, 0 1, (ξ, η) ∈ Λ × Λ. 4) where c is a positive constant independent of h. Proof. Let e i = u (xi ) − u i and e i = v (xi ) − v i . 5) where ξi and ξi are two intermediate values between u (xi ) and u i . Deﬁne T E = (e 1 , e 2 , .

Comp. 11 (1957) 257–261. A. W. G. J. McRae, An eﬃcient method for parametric uncertainty analysis of numerical geophysical model, J. Geophys. Res. 102 (1997) 21925–21932. [18] A. Taﬂove, Computational Electrodynamics – The Finite-Difference Time-Domain Method, Aztech House, Boston, 1995. C. N. Trefethen, The kink phenomenon in Fejer and Clenshaw–Curtis quadrature, Report no. 6/16, Oxford University Computing Laboratory, 2006. [20] N. Wiener, The homogeneous chaos, Amer. J. Math. 60 (1938) 897–936.