Jump to content
GMS, SMS, and WMS User Forum
Sign in to follow this  
hycc_sms

How to simulate a reflect wall in Bouss-2D?

Recommended Posts

In Bouss2D, If simulate wave passing a wall which is 150m long and 20m wide, then how to simulate the reflective wave by the wall?

I set a porosity layer around the wall, the layer width is 4m,and the value is 0(full reflect). But the computation is interrupt at the beginning of the simulation.

So My quesion is how to set the boundary condition to simulate the reflect of wall? How to determine the coefficient(width, value, etc,.)

Plz tell my as soon as possible.

Thanks!

Share this post


Link to post
Share on other sites

cause the wall is vertical, so the elevation is suddenly from -10m to 2m across one grid,

so the computation is interrupt in the wall area.

How to deal with thia case?

Thanks!

Share this post


Link to post
Share on other sites

When using CGWAVE, the perimeter of the domain is divided into nodestrings to represent the boundary conditions for CGWAVE. Two types of boundary conditions exist including open ocean and coastline. Each section of coastline and the ocean interface are stored in SMS as nodestrings. Each coastline section can be assigned its own reflection coefficient to tell CGWAVE how waves will reflect from that portion of shoreline. Values for reflection coefficients vary between 0.0 and 1.0 and typical values include:

Reflection Coefficient      Shoreline Condition
     1.0                       Vertical wall
     0.4-0.5                   Large Rocks
     0.15-0.25                 Sandy Beach

To assign initial values for reflection coefficients:

  1. Click on the Select Nodestring tool. Nodestring handles (boxes) will appear over the nodestrings in the model. The coastline nodestrings are colored brown. The blue nodestrings represents the ocean.
  2. Select the coastline nodestrings you want to assign reflection coefficients to by dragging a box around them.
  3. Select the CGWave | Assign BC command. Make sure that Coastline is selected and set the Reflection Coefficient to 1.0 (wall).
  4. Click OK.

It is important to note that to model a vertical or steep wall in CGWAVE, you should use the reflection coefficients, rather than representing the wall as part of your mesh. A steep vertical element will cause the model to be unstable. The vertical wall should be modeled using appropriate reflection coefficients, but not represented in the mesh by elevation change.

You may want to review the test cases posted on the xmswiki, in particular:

Test 18

Propagation of obliquely incident waves (incidence angle =20°) past a shore-perpendicular thin fully-reflecting breakwater on a sloping beach (beach is fully absorbing). Parabolic and one-dimensional open boundary condition. Results match analytical results given in Kirby (1986) and Panchang et al. (2000).

Tests 19 & 20

As in Test 18, but with nonlinear breaking on and off. Test pertains parameters in Zhao et al., (2000, Coastal Engineering).

Test 21

As in Test 19, but with shore-parallel breakwater. Parameters and results as in Zhao et al. (2000). Results are for no breaking.

Test 22

As in Test 19, but with shore-parallel breakwater. Parameters and results as in Zhao et al. (2000). Results are for nonlinear breaking.

Test 23

Wave propagation/resonance in a rectangular harbor. Results match analytical solution plotted in Demirbilek & Panchang (1998). With friction f = 0.12, the resonant peak amplification reduces substantially for kl =1.4 (T = 1.0447 s).

Share this post


Link to post
Share on other sites

Sorry about that! I had just answered a different CGWAVE question and had CGWAVE on the mind I guess.

You said in your original post you were using a porosity boundary condition - are you modeling a porous structure? Your second post seems to indicate that it is a vertical structure?

Remember that in BOUSS2D, all interfaces between ocean and land cells and all ocean boundaries are considered vertical walls and will reflect waves (like the walls of a physical model in a wave tank). If you have a vertical wall, you can represent it by using land cells. Damping can be used to add energy loss to waves interacting with the structure, but is not required.

One problem that is frequently encountered with highly reflective structures is having reflected waves bounce back into the wavemaker. They will then hit the boundary behind the wavemaker and bounce back into the domain. If you are getting waves reflecting back into your wavemaker, you can either:

1. Try putting your wavemaker on the boundary and further away from the source of reflections so the reflected waves die out before they reach the wavemaker

2. Move the wavemaker in from the boundary and add damping cells behind the wavemaker to absorb the waves generated by the wavemaker and the reflected waves

Figure 5 on page 24 of the BOUSS2D manual provides some insight into the variation of reflection coefficient with damping coefficient. The values used for width and coefficient value are highly dependent on the situation. When using damping to represent dissipation of energy due to a structure, you will want to do a sensitivity analysis or calibrate to real data. Also, remember that increasing the width of a damping boundary condition causes your structure to influence a larger area away from the structure.

Share this post


Link to post
Share on other sites

Thanks!

I set the vertical wall land value for 2.0m, and water depth all arround is 9.0m. When run this simulation , the SMS then warning that " slope exceeds max recommended value of 0.5 at I=..,J-.." BUt there is no slope in this case now, why? And dose this effect the simulation stability?

Regards!

Share this post


Link to post
Share on other sites

If you set the land value to 2.0 m and the adjacent cell to 9.0 m, then over the distance of one cell you change elevation by 7.0 m. If you turn on contours or rotate the grid and view it from a side angle, you should be able to see this. You can also turn on the "computational domain" in the display options to see areas that might be problematic.

You are correct in stating that BOUSS2D is more likely to become unstabel with steep slopes. For example, on page 36 of the manual it notes that for the Waver Runup Option - "The runup scheme used in the numerical model is designed to simulate subcritical flow conditions on mild slopes. It cannot resolve details of supercritical flow on steep slopes."

Share this post


Link to post
Share on other sites

Sorry, I miswrite.I set land value 2.0, and water value -9.0m.

Yes, if compute runup, I must make a mild slope in the area, but now I don't make the runup option on,I just want to simulate wall reflection.

And clearly, the wall must be land value, and be surrounded by water. So it seems that this warning is inevitable in wall simulation,right?

Thanks!

If you set the land value to 2.0 m and the adjacent cell to 9.0 m, then over the distance of one cell you change elevation by 7.0 m. If you turn on contours or rotate the grid and view it from a side angle, you should be able to see this. You can also turn on the "computational domain" in the display options to see areas that might be problematic.

You are correct in stating that BOUSS2D is more likely to become unstabel with steep slopes. For example, on page 36 of the manual it notes that for the Waver Runup Option - "The runup scheme used in the numerical model is designed to simulate subcritical flow conditions on mild slopes. It cannot resolve details of supercritical flow on steep slopes."

Share this post


Link to post
Share on other sites

Is the wall you are trying to simulate a vertical wall? BOUSS-2D treats all land / water boundaries as vertical walls (just like the sides of a wave tank), so to model a perfectly vertical wall all you need is a land/water boundary. You don't need to represent the vertical wall by changing the elevations.

You can manually assign a cell to be land (without changing the elevation) by selecting the cell and using the menu "BOUSS-2D | Assign Cell Attributes." A dialog will open in which you can specify the cell type as ocean or land.

Share this post


Link to post
Share on other sites

I do a simple case and set the land cell as you say. Yes, this can simulate a vertical wall in water, Thanks!

And I have another question that, if simulate a steep slope structure, e.g. a rubble mound breakwater, then how should I do?

should I make a steep slope change in elevation in the structure area, or I should only set land cell for crest of breakwater and then set a damping layer around it?

Regards!

Share this post


Link to post
Share on other sites

And I have another question about current inBouss2D.

If I want to compute the wave propagation in steady current velocity, how to set the current in SMS?

I saw a drop-down menu"Current", but there is only "None" in this menu.

Share this post


Link to post
Share on other sites

The reason only "None" appears in the "Input Data Sets - Current" section of the BOUSS-2D Model Control is there are no vector data sets associated with the cartesian grid. To use currents with BOUSS-2D, either create a data set using the data calculator, interpolate the results from say an ADCIRC run or some other model to the cartesian grid, etc. Once a vector data set is associated with the cartesian grid, the vector data set's name will appear in the drop-down menu.

Share this post


Link to post
Share on other sites
To simulate a steep slope structure, e.g. a rubble mound breakwater, then how should I do? Should I make a steep slope change in elevation in the structure area, or I should only set land cell for crest of breakwater and then set a damping layer around it?

To simulate a sloped structure, you would represent the structure by a change in elevation in the structure area.

The damping layer is used to add any dissipation of energy that occurs when the waves hit the structure, which will reduce the amount of reflection.

Share this post


Link to post
Share on other sites

But if make these elevation to represent structure, the steep slope will exceed the max recommended slope value, the computation then will be unstable, what should I do for that ?

To simulate a sloped structure, you would represent the structure by a change in elevation in the structure area.

The damping layer is used to add any dissipation of energy that occurs when the waves hit the structure, which will reduce the amount of reflection.

Share this post


Link to post
Share on other sites

The only other option that I can think of to representing a structure by way of the bathymetry is to use Porosity layers.

Page 30 of the manual states, "Porosity layers are used to simulate partial reflection and transmission through porous structures such as breakwaters. A porosity value of 0.4 is typically used for breakwaters. Regions with small porosity values (n < 0.1) such as breakwater core layers should be treated as impermeable regions and mapped as land points in the bathymetry file."

Share this post


Link to post
Share on other sites

The only other option that I can think of to representing a structure by way of the bathymetry is to use Porosity layers.

Page 30 of the manual states, "Porosity layers are used to simulate partial reflection and transmission through porous structures such as breakwaters. A porosity value of 0.4 is typically used for breakwaters. Regions with small porosity values (n < 0.1) such as breakwater core layers should be treated as impermeable regions and mapped as land points in the bathymetry file."

Dear both:

I am fully interested on your discussion, since I have a similar problem. I would like to add one more comment: I am using 10.0 as well. The run is O.K. with dampiung only, but if I change structures (cliffs, rubble mound breakwaters and vertical quays) into porosity layers, to account for reflections, I inmediately got this error (please, see the attached file). Sop, I am using only damping layers 3 cells width.

One more comment: Is there any utility to perform spectral analysis from the timeseries of WSE? I have my own system through matlab, but if possible doing with SMS, I would prefer.

Cheers,

Jose

post-969-125222937091_thumb.png

Share this post


Link to post
Share on other sites

Hello everyone,

I've been having the same problem, and I think I could help with this discussion.

First of all, despite all the good things that bouss2d has, it's worth to know that theirs equations of motion CAN'T represent phenomenons such like of propagations over breakwaters, since its particular set of Boussinesq-type equations is not fully non-linear and overall it assumes in their derivation that the bottom has a mild slope.

With this in mind, you have to know that any alternative to simulate a breakwater or any similar structure, like using damping layers or similar, will be always and "ad-hoc" approximation and it's not warranty to work successfully in all cases.

You can here find an excellent paper where this is remarked

http://www.dma.ens.fr/~lannes/fichiers/Lannes-Bonneton_PhysFluids.pdf

Finally, I'd like to say that despite bouss2d is not the best alternative to model such kind of cases, it's one of the few available commercial alternatives for doing so, without dealing with very academic models, so nevertheless their results will be approximated, they are a good approximation.

Cheers,

Benjamín Carrión

Hydraulic Engineer

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
Sign in to follow this  

×
×
  • Create New...