Numerical techniques for flow problems with singularities.

*(English)*Zbl 1032.76637Summary: This paper deals with grid approximations to Prandtl’s boundary value problem for boundary layer equations on a flat plate in a region including the boundary layer, but outside a neighbourhood of its leading edge. The perturbation parameter \(\varepsilon = Re^{-1}\) takes any values from the half-interval (0,1] ; here Re is the Reynolds number. To demonstrate our numerical techniques we consider the case of the self-similar solution. By using piecewise uniform meshes, which are refined in a neighbourhood of the parabolic boundary layer, we construct a finite difference scheme that converges \(\varepsilon\)-uniformly. We present the technique of experimental substantiation of \(\varepsilon\)-uniform convergence forboth the numerical solution and its normalized (scaled) difference derivatives, outside a neighbourhood of the leading edge of the plate. By numerical experiments we demonstrate the efficiency of numerical techniques based on the fitted mesh method. We discuss also the applicability of fitted operator methods for the numerical approximation of the Prandtl problem. It is shown that the use of meshes refined in the parabolic boundary layer region is necessary for achieving \(\varepsilon\)-uniform convergence.

##### MSC:

76M20 | Finite difference methods applied to problems in fluid mechanics |

76D10 | Boundary-layer theory, separation and reattachment, higher-order effects |

65N15 | Error bounds for boundary value problems involving PDEs |

65N12 | Stability and convergence of numerical methods for boundary value problems involving PDEs |

##### Keywords:

boundary-layer equations; difference schemes; piecewise-uniform fitted mesh; fitted operator method; \(\varepsilon\)-uniform convergences
PDF
BibTeX
XML
Cite

\textit{P. A. Farrell} et al., Int. J. Numer. Methods Fluids 43, No. 8, 915--936 (2003; Zbl 1032.76637)

Full Text:
DOI

**OpenURL**

##### References:

[1] | Boundary Layer Theory (7th edn). McGraw Hill: New York, 1979. |

[2] | Mathematical Models in Boundary Layer Theory. Chapman & Hall/CRC Press: Boca Raton, FL, 1999. |

[3] | Methods of Numerical Mathematics. Springer: New York, 1982. |

[4] | Theory of Difference Schemes (3rd edn). Nauka: Moscow, 1989 (English transl.: The Theoryof Difference Schemes. Marcel Dekker, Inc.: New York, 2001) (in Russian). |

[5] | Shishkin, USSR Computational Mathematics and Mathematical Physics 32 pp 467– (1992) |

[6] | Farrell, Mathematics of Computation 67 pp 603– (1998) |

[7] | Shishkin, USSR Computational Mathematics and Mathematical Physics 29 pp 1– (1989) |

[8] | Grid Approximations of Singularly Perturbed Elliptic and Parabolic Equations. Ural Branch of Russian Academy of Sciences: Ekaterinburg, 1992 (in Russian). · Zbl 1397.65005 |

[9] | Fitted Numerical Methods for Singular Perturbation Problems. World Scientific: Singapore, 1996. |

[10] | Numerical Methods for Singularly Perturbed Differential Equations. Convection-Diffusion and Flow Problems. Springer: Berlin, 1996. |

[11] | Shishkin, USSR Computational Mathematics and Mathematical Physics 31 pp 53– (1991) |

[12] | Hegarty, CWI Quarterly 10 pp 239– (1997) |

[13] | Benchmark computations of laminar flow around a cylinder. In Flow Simulation with High-Performance Computation II, (ed). Notes on Numerical Fluid Mechanics, vol. 52. Vieweg-Verlag: Braunschweig, 1996; 547-566. · Zbl 0874.76070 |

[14] | Shishkin, Computational Mathematics and Mathematical Physics 41 pp 37– (2001) |

[15] | Hegarty, Journal of Computational Physics 117 pp 47– (1995) |

[16] | Robust Computational Techniques for Boundary Layers. Chapman & Hall/CRC Press: Boca Raton, FL, 2000. · Zbl 0964.65083 |

[17] | Liao, International Journal of Non-Linear Mechanics 34 pp 759– (1999) |

[18] | Shishkin, Soviet Mathematics Doklady 37 pp 792– (1988) |

[19] | Hemker, Computational Fluid Dynamics Journal 2 pp 375– (1994) |

[20] | Shishkin, Journal of Mathematical Analysis and Applications 208 pp 181– (1997) |

This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.