Loading...
Please wait, while we are loading the content...
Similar Documents
High-order DNS and LES simulations using an implicit tensor-product discontinuous Galerkin method
| Content Provider | Semantic Scholar |
|---|---|
| Copyright Year | 2017 |
| Abstract | This paper describes an efficient tensor-product based preconditioner for the large linear systems arising from the implicit time integration of discontinuous Galerkin (DG) discretizations. A main advantage of the DG method is its potential for high-order accuracy, but the number of degrees of freedom per element scales as p, where p is the polynomial degree and d is the spatial dimension. Standard preconditioners such as block Jacobi and ILU factorizations rely on dense linear algebra, incurring a computational cost of O(p), which quickly becomes intractable. Our new preconditioner exploits the natural tensor-product structure of general quadrilateral and hexahedral meshes to reduce the computational complexity to O(p) in two dimensions, and O(p) in three dimensions. We apply this preconditioner to two benchmark fluid flow problems: the direct numerical solution of the compressible Taylor-Green vortex, and the large eddy simulation of flow over a NACA airfoil. These test cases demonstrate the effectiveness of the new tensor-product preconditioners on large-scale, high-order CFD problems. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://persson.berkeley.edu/pub/pazner17implicitdns.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
