|
[Sponsors] |
November 23, 1999, 04:24 |
MULTIBLOCK APPROACH
|
#1 |
Guest
Posts: n/a
|
DEAR SIR I AM INTERESTED TO KNOW MORE ABOUT MULTIBLOCK APPROACH AS A MESH GENERATIG TOOL AND ALSO AS A SOLVING METHOD, AND IT'S ADVANTAGES.IS IT SUITABLE FOR 3-D MODELLIG OF FLOW IN THE DRAFT TUBE OF TURBINE? PLEASE DIRECT ME.
THANK YOU |
|
November 25, 1999, 18:25 |
Re: MULTIBLOCK APPROACH
|
#2 |
Guest
Posts: n/a
|
(1). Structured multi-block mesh and the related solver basically is the same as the single block mesh, except in the booking of the individual block. (2). For example, a single block mesh of 60x40 can be divided into two blocks as 30x40 and 30x40. And the mesh can be created separately as if each is a single block mesh. (3). Since in this simple example, there is no internal boundary at i=30, the solver has to handle it as interior points, so that solutions can communicate with each block. That is the principle. (4). To eliminate this extra step, one can add extra mesh line from the joining mesh to the other mesh, to create overlapping region. In this way, the solution of the first block will include an extra mesh line solution, which will act as the boundary condition for the next block solution. You can use a couple of paper cards, with one edge of the card on the top of the other one, to give you a better picture. (5). If you slide one card upward such that only portion of the edge is overlapped, then it will represent a general multi-block mesh approach. (6). Using this principle, you can attach more cards (meshes) to the original cards, to create more complex multi-block mesh. (7). As you can see that, there is no extra cards ( or meshes) wasted (blank un-used). Thus, the computer memory is used effectively. (8). If one is not concerned about the memory, the structured multi-block mesh can be replaced be a conventional single block mesh. In this case, the solver avoids the region outside the internal boundary and the block-out region is not solved. (similar to a house with several locked un-used rooms) (9). So, for structured multi-block mesh and solver, it is fairly simple. You simply solve the multi-block mesh, block by block, each with an extra mesh line so that it will act as the boundary condition of the next block.
|
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Lattice Boltzmann Approach | philippose | OpenFOAM | 20 | December 3, 2016 08:39 |
Multiblock ctructured grids in CGNS-format | Beaver | Main CFD Forum | 0 | April 4, 2010 07:58 |
whether DEM or Eulerian approach for high solidup | CFDUSERIN | CFX | 1 | April 19, 2007 10:18 |
Data structure handling in multiblock approach | CH | Main CFD Forum | 3 | October 5, 2006 19:51 |
lagrangian vs eulerian approach | leopoldo luongo | CFX | 1 | June 19, 2006 11:12 |