arXiv · 2611.16959
An algorithm for solving the Burgers equation based on rational approximations and the collocation method
Abstract
We propose a numerical method for solving initial-boundary value problems for the Burgers equation that takes into account the presence of singularities of the analytic continuation of the solution into the complex plane. The algorithm uses rational approximations, a collocation method and a new scheme for time discretization combined with the relaxation method. An adaptive strategy is developed to account for the location of the singular point of the solution. Solutions to the well-known benchmark problem with a sine-shaped initial condition are obtained for the values of the viscosity as low as nu = 10^-8. Verification of the numerical results in such an extreme parameter regime makes it necessary to modify a classical formula for the exact solution so as to avoid floating-point overflow during its computation. The developed method is compared with an algorithm employing the Runge-Kutta scheme, and the conditions under which each approach yields superior results are outlined. An important feature of the proposed method is its capability to track the singular point of the solution. To verify this, we consider a problem whose solution has a simple analytical form and demonstrate that the exact and approximate coordinates of the singular point of the solution agree with high accuracy, even when the singularity comes extremely close to the domain of the problem. Taking the location of the singular point into account yields significant gains in both accuracy and run time relative to other methods.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
I. A. Bugoets, B. V. Semisalov. 2026-11-12. An algorithm for solving the Burgers equation based on rational approximations and the collocation method. https://arxiv.science/abs/2611.16959
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite arXiv Science by Indiciae: BibTeX · RIS.