Scribd
Upload
142 views2 pages

Lecture1 8 PDF

The document discusses grid generation for computational fluid dynamics (CFD) simulations. It outlines the two main types of grids - structured and unstructured - and describes their characteristics. For complex geometries, a block-structured or chimera approach can be used where the domain is divided into blocks which can each have a different grid structure. The grid generation process typically involves decomposing the domain, generating edge and surface grids, and then using these to create the volume mesh. Grid generation is an important part of CFD that significantly impacts solution accuracy.

Uploaded by

biruk1
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
142 views2 pages

Lecture1 8 PDF

The document discusses grid generation for computational fluid dynamics (CFD) simulations. It outlines the two main types of grids - structured and unstructured - and describes their characteristics. For complex geometries, a block-structured or chimera approach can be used where the domain is divided into blocks which can each have a different grid structure. The grid generation process typically involves decomposing the domain, generating edge and surface grids, and then using these to create the volume mesh. Grid generation is an important part of CFD that significantly impacts solution accuracy.

Uploaded by

biruk1
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

COMPUTATIONAL FLUID DYNAMICS: GRID GENERATION PROESS

Lecture 30
TYPES OF GRIDS AND GRID GENERATION PROCESS

30.1 INTRODUCTION
A mesh or grid is a set of points distributed over the problem domain for a numerical solution
of a set of partial differential equations (PDEs). In CFD analysis, type of grid/mesh would
depend on the discretization technique (FDM/FVM/FEM), geometry of the problem domain
and underlying physics. Structured grid is required for the finite difference method, whereas
FEM and FVM can work with either structured or unstructured grids. In case of unstructured
grids, care must be taken to ensure proper grading and quality of the mesh.

The grid/mesh strongly affects the accuracy of the numerical solution, and hence, significant
effort is usually invested in generation of a suitable grid. In industrial CFD analyses which
involve numerical solution of problems in complex geometries, mesh generation process may
consume nearly 50% of overall simulation time.

A lot of research effort has been invested in automatic (numerical) grid generation in past few
decades, and it is still an active research area. In this introductory course on CFD, it is not
possible to take a detailed look at different mesh generation techniques. Hence, this and next
lecture would primarily provide a descriptive outline of automatic grid generation. Interested
reader can refer to suggested books/references for complete mathematical details of different
algorithms and their computer implementation.

30.2 TYPES OF GRIDS


Majority of grids can be broadly put into two categories:
1. Structured grids
2. Un-structured grids

Structured Grids
In structured grids, grid points follow a fixed structure. In particular,
 Grid points are located at the intersections of the grid lines. Hence, each grid points
can represent the origin of a local coordinate system.
 Interior grid points have a fixed number of neighboring points.
 Grid points can be represented by a set of indices corresponding to the intersecting
grid lines.
Structured grids consists of elements which can be mapped to a rectangle (in 2-D) or
parallelepiped (in 3-D). Thus, all grid lines are oriented regularly in either two or three
directions so that co-ordinate transformation of curvilinear lines results in a rectangle in 2D
and parallelepiped in 3D.

Structured grids can be Cartesian or body-fitted grids. The former are used in simple
rectangular geometries, whereas the latter are preferred for domains with curved boundaries.
Further, the body-fitted grids can be
(a) orthogonal (in which curvilinear grid lines are perpendicular at the grid-points), or

Dr K M Singh, Indian Institute of Technology Roorkee NPTEL 30.1


COMPUTATIONAL FLUID DYNAMICS: GRID GENERATION PROESS

(b) non-orthogonal (in which curvilinear grid lines intersect obliquely at the grid-points).

Block-structured Grids
For complex geometries, generation of a body-fitted structured grid may be very difficult. In
this case, it is usually advisable to decompose the problem geometry in a set of blocks (sub-
domains), and generate a structured grid in each in each block separately. The grid structure
in each block can be different.

Block-structured grids allow different mesh densities in different regions (blocks), and thus
fine grids can be easily put in regions likely to show steeper gradients of problem variables.
The blocks can even be overlapping. Such block-structured grids (i.e. grids with overlapping
subdomains) are called composite or chimera grids.

Un-structured Grids
Unlike structured grids, sides of a cell/element have no relation to the co-ordinate directions
in unstructured grids. The un-structured grid consists of triangles or quadrilateral in 2D and
tetrahedra, wedge, hexahedra or polyhedra in 3D.

Unstructured grids are much easier to generate in complex domains, and hence, majority of
commercial CFD solvers have adopted unstructured grid based solvers. In finite volume
applications, quadrilateral (hexahedral) elements are preferred for better accuracy in
interpolation and integrations.

30.2 GRID GENERATION PROCESS


The process of grid-generation for complex geometries normally involves following steps:
 Decompose the problem domain into a set of sub-domains (blocks)
 In each block, generate the requisite grid. Typical sequence of operations would be
o Generate edge-grid (i.e. divide the edges of a surface in desired number of one
dimensional elements).
o Using the edge grids, generate the grid on the block-surfaces.
o Use surface grids as input to generate volume mesh.
 Check mesh quality, and modify the mesh as required.

REFERENCES/FURTHER READING
Chung, T. J. (2010). Computational Fluid Dynamics. 2nd Ed., Cambridge University Press.
Liseikin, V. D. (2006). A Computational Differential Geometry Approach to Grid
Generation. Springer.
Taniguchi, T. (2006). Automatic Mesh Generation for 3D FEM, Robust Delaunay
Triangulation. Morikita Publishing.
Thompson, J. F., Soni, B., Weatherill, N. P. (1999). Handbook of Grid Generation. CRC
Press.
Thompson, J.F., Warsi, Z.U.A., Mastin, C. W. (1985). Numerical Grid Generation,
Foundations and Applications. North Holland.

Dr K M Singh, Indian Institute of Technology Roorkee NPTEL 30.2

You might also like