The example has an extremely coarse grid, which makes the computational time relatively short. However, this also means that the flow field is not resolved properly around the structure, and the scour depth becomes much smaller than it should be.
Some comments about the data set in the control file for the example:
1. The F 7 B option is used. This is an important parameter in reducing the critial shear stress of the particles as a function of the sloping bed. The algorithm has three empirical parameters, given on the F 109 data set. These parameters will affect the shape and depth of the scour hole. Instead of using F 7 B, the F 182 data set can be used.
2. The F 16 option is used to determine the bed roughness. The value of this data set should be equal to the effective roughness of the particles on the bed, for example 3 times d90.
3. There are far too few inner iterations specified on the F 33 data set (second parameter). A typical value to get good results is about 100-1000.
4. The transient free surface algorithm is used (F 36 2). This causes increased computational time. If the water surface is fairly horizontal so that its location does not affect the flow field significantly, the computational time can be shortened by removing this data set.
5. The F 56 parameter invokes a sand slide algorithm. The second parameter on this data set will affect the shape and magnitude of the scour hole. The F 56 200 option seems to give the best results.
6. If you want to model circular cylinders with a nested grid using SSIIM 2, then the following file will be useful: Download 'koosurf.nes' file
|
|
|
|
