scientific article; zbMATH DE number 3965715

From MaRDI portal
Publication:3734080





[For the entire collection see Zbl 0562.00012.] Finite element approximation theory is employed to develop a numerical solution algorithm for the three-dimensional Navier-Stokes equations governing inviscid, viscous and/or turbulent flow of a compressible fluid. The fundamental theoretical statement requires the semi-discrete approximation error to be orthogonal to the space of functions employed in forming the approximation. Various additional constraints can be appended to the basic statement as necessary. For example, the additional constraint requiring the semi-discrete approximation error in the substantial time derivative terms of the Navier-Stokes equations to be orthogonal to a modified function space yields a highly phase selective artificial dissipation mechanism for control of non-linearly induced instabilities. An idealized von Neumann analysis is employed to estimate the functional form of the constraint, and numerical experiments for shocked ducted flows are employed to refine these estimates. Results of several basic computational experiments confirm additional theoretical and practical aspects of the algorithm.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3734080)