This article gives a historical perspective on the evolution of implicit algorithms for the Navier-Stokes equations that utilize time or Newton linearizations and various forms of approximate factorization including alternating-direction, lower/upper, and symmetric relaxation schemes. A theme of the paper is how progress in implicit Navier-Stokes algorithms has been influenced and enabled by the introduction of characteristic-based upwind approximations, unstructured-grid discretizations, parallelization, and by advances in computer performance and architecture. Historical examples of runtime, problem size, and estimated cost are given for actual and hypothetical Navier-Stokes flow cases from the past 40 years. (C) 2010 Elsevier Ltd. All rights reserved.
机构:
Morgan State Univ, Dept Math, 1700 E Cold Spring Ln, Baltimore, MD 21251 USAMorgan State Univ, Dept Math, 1700 E Cold Spring Ln, Baltimore, MD 21251 USA
Cai, Mingchao
Huang, Peiqi
论文数: 0引用数: 0
h-index: 0
机构:
Nanjing Forestry Univ, Dept Appl Math, Nanjing 210037, Jiangsu, Peoples R ChinaMorgan State Univ, Dept Math, 1700 E Cold Spring Ln, Baltimore, MD 21251 USA
Huang, Peiqi
Mu, Mo
论文数: 0引用数: 0
h-index: 0
机构:
Hong Kong Univ Sci Technol, Dept Math, Kowloon, Hong Kong, Peoples R ChinaMorgan State Univ, Dept Math, 1700 E Cold Spring Ln, Baltimore, MD 21251 USA