An Introduction to the Numerical Analysis of Spectral by Bertrand Mercier

By Bertrand Mercier

Show description

Read or Download An Introduction to the Numerical Analysis of Spectral Methods PDF

Similar computational mathematicsematics books

Analytical and numerical approaches to asymptotic problems in analysis: proceedings of the Conference on Analytical and Numerical approaches to Asymptotic Problems, University of Nijmegen, the Netherlands, June 9-13, 1980

A world convention on Analytical and Numerical ways to Asymptotic difficulties used to be held within the school of technology, collage of Nijmegen, The Netherlands from June ninth via June thirteenth, 1980.

Applied Stochastic Processes and Control for Jump-Diffusions: Modeling, Analysis, and Computation (Advances in Design and Control)

This self-contained, useful, entry-level textual content integrates the elemental rules of utilized arithmetic, utilized chance, and computational technology for a transparent presentation of stochastic approaches and keep an eye on for jump-diffusions in non-stop time. the writer covers the real challenge of controlling those platforms and, by using a leap calculus building, discusses the robust function of discontinuous and nonsmooth houses as opposed to random houses in stochastic structures.

Computational Science – ICCS 2007: 7th International Conference, Beijing, China, May 27 - 30, 2007, Proceedings, Part III

A part of a four-volume set, this publication constitutes the refereed lawsuits of the seventh foreign convention on Computational technological know-how, ICCS 2007, held in Beijing, China in may possibly 2007. The papers disguise a wide quantity of subject matters in computational technology and similar parts, from multiscale physics to instant networks, and from graph idea to instruments for software improvement.

Additional info for An Introduction to the Numerical Analysis of Spectral Methods

Example text

1) u(x,0) = u0(x) where u 0 e D(L) is given. 1) admits u ~ C0(0,r;H~(1)). ,t)ll < Cllu01Ls, for t s [0,T], S where T is positive and given. Proof: The proof of this result is an elegant applieaton of the theory of pseudo-differential operators (see M. Taylor, [17], pp. 62-65). content ourselves with establishing the a priori estimate in solution assuming it exists. 1) we have then, by setting K = [AS,L] --- ASL - LA s d 2 d 2 ~u ASu) + (hSu, A s ~u d'-~ llu(t)l]s = d'~ t]hSu(t)110 = (AS ~ ' ~-t) = - (ASLu, ASu) - (ASu,ASLu) = - (LESu, hSu) - (Ku, hSu) - (hSu ,LASu) ~ (hSu, Ku) = - (Ku, ASu) - (ASu, Ku), where we have utilized the antisymmetry of Since order K L.

Be the solution of the approximate adjoint problem aW N , (~--~--+ L WN,VN) = 0, for all v N ~ SN (WN(O) - ¢,VN) = 0, for all v N e SN. 2, if ~ e lip (I), lIW(t) - WN(t)li 0 < C N-Sll~lls+I for t < T. 7) (¢,UN(t) - u(t)) < C N-SH~Hs+itlu0fl 0. 7) as an error estimate in the Sobolev space of negative indices. 7) converges to zero as ~ is regular). This explains intuitively the success of the Fourier method with smooth- 41 ing, consisting Osher, [12]). of smoothing the initial solution By that we mean the following; let u 0 (see Majda-McDonoughp be a positive regular function with a compact support such that: / p(x)dx = I.

Furthermore, from the definition of the norm li. 1) (applied with r = T'I) IIY-PcYll0 < C(I+N2)-(T-I)/211ylIT_I C(l+N2)-(T-l)/2tiull . that 60 We have II(Lc-L)]NIi 0 < II(Pc-I)yll0 + H(Pc-I)a]NII I C(l+N2)(l-~)/2iluIIT + ;i(Pc-l)a]Nil I. 5) we have liauNiIy < CIIUNII~ This establishes ilauNflT, Cil uil T. 6) ilu(t)ll . 8) z ~u )--f- ~u 11~_~ii0 = il(l_PN) ~-t li0 ~ C(I+N 2 U~IIT-I < C(I+N 2) 2 l-r i-~ liullr' 61 where the last inequality is gotten by noting ~U B--~ = -Lu, IIL u II~ - 1 < Cllul} .

Download PDF sample

Rated 4.09 of 5 – based on 11 votes