Issue |
ESAIM: M2AN
Volume 46, Number 6, November-December 2012
|
|
---|---|---|
Page(s) | 1407 - 1420 | |
DOI | https://doi.org/10.1051/m2an/2012004 | |
Published online | 19 April 2012 |
Efficient computation of delay differential equations with highly oscillatory terms
1
School of Electronic Engineering, Dublin City
University, Dublin 9,
Ireland
marissa.condon@dcu.ie
2
Dpto. de Matemáticas, Universidad Carlos III de Madrid, Avda.
Universidad, 30,
Leganés
28911, Madrid, Spain
3
Department of Applied Mathematics and Theoretical Physics, Centre
for Mathematical Sciences, University of Cambridge, Wilberforce Rd, CB3 0 WA Cambridge, UK
4
Institute of Mathematics, University of Gdańsk,
Wit Stwosz Str. 57,
80-952
Gdańsk,
Poland
Received: 18 March 2011
Revised: 15 December 2011
This paper is concerned with the asymptotic expansion and numerical solution of systems of linear delay differential equations with highly oscillatory forcing terms. The computation of such problems using standard numerical methods is exceedingly slow and inefficient, indeed standard software is practically useless for this purpose. We propose an alternative, consisting of an asymptotic expansion of the solution, where each term can be derived either by recursion or by solving a non-oscillatory problem. This leads to methods which, counter-intuitively to those developed according to standard numerical reasoning, exhibit improved performance with growing frequency of oscillation.
Mathematics Subject Classification: 34E05 / 34E99 / 42A99 / 34K28
Key words: Delay differential equations / asymptotic expansions / modulated Fourier expansions / numerical analysis
© EDP Sciences, SMAI, 2012
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.