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

Access violation error on BOUSS2D

Recommended Posts

Dear all,

I've been working with Bouss2d, and I started to been notified with this error:

"forrtl: severe (157): Program Exception - access violation"

As I've found on internet, this might have relation with the program writing in pieces of memory where it is not allowed to write, or with the change of the file's paths.

In my opinion I haven't done any of this things, but now I keep receiving this message and I cannot finish my models. Please, any help or leads on why this could be happening will be very appreciated.

Thanks in advance,

Benjamín

error.bmp

Share this post


Link to post
Share on other sites

Benjamín,

What version of SMS are you using? The BOUSS-2D model executable for SMS 10.1 is not currently working. We are working with the model developer to find a solution. This is a problem with the model and not SMS. You can either work in SMS 10.0 or use the BOUSS-2D model executable from SMS 10.0 with SMS 10.1. http://xmswiki.com/xms/SMS:Bugfixes_SMS_10.1

Share this post


Link to post
Share on other sites

Benjamín,

I just noticed the image you posted. I see the model is giving slope warnings.

http://xmswiki.com/xms/SMS:BOUSS-2D#Instability_due_to_the_grid.2Fgeometry

"Abrupt changes in elevation from one cell to another in the computational domain could result in instabilities. You may find it helpful to smooth the grid. (A smoothing command is available by right clicking on the grid object in the project explorer in the SMS interface.)"

BOUSS-2D Manual, page 35:

"For plunging waves on steep slopes, it may be necessary to use up to 100 points per wave period to simulate the rapid changes in wave shape irrespective of the Courant number"

Some general thoughts:

  • Steep slopes (1:1 or 1:2) near the shore can be represented by a vertical wall with damping.
  • If the depth is greater than the wavelength / 2 you may want to truncate the bathymetry to a flat bottom. This will help the accuracy of the wave solution while sacrificing the accuracy of the circulation in the deep region.
  • Structures should be at least two to three wavelengths (up to ten) away from the boundary so that reflections don't contaminate the boundary.
  • If the abrupt change in slope is a transition from deep to shallow water, you may want to use the smooth data set tool to smooth the bathymetry to keep reflections from contaminating the boundary. Anchor the maximum value and select an appropriate maximum slope.

Share this post


Link to post
Share on other sites

John,

Thanks for your reply and advices. I'm working with the 10.0 version and it's working fine... for most cases.

From your ideas, I'm inclined to think that the problem might have relation with the breakwater that I'm trying to model (http://aquaveo.invisionzone.com/index.php?showtopic=699) since I have represented it as part of the bathymetry and, in fact, it presents slopes in the order of 1:1.75.

Nevertheless, I'm using a grid discretization (dx=dy=5m) smaller that the one recommended by the model, and smaller time steps, to achieve Courant's Numbers always below 0.5.

In any case, the model works well for waves with relative shorter periods, up to 12 seconds, and begins to fail from 15 seconds and above, when the cells per wave begin to be fewer, and therefore the model might be unable to reproduce the fast shoaling process, as you suggest.

Anyway, I see three possible attempts for solving this:

- Using a smaller dx, since it worked well for shorter waves

- Try to smooth the scattered data... but I'm not sure if this is a good idea, since there are elevation levels that must be maintained to represent the breakwater

- Or else, try to represent the breakwater in another way, as a vertical wall like you say, but in that case I'm afraid to alter the output by a bad choice of the damping parameters. In other words, despite the problems that the model might have to deal with large bathymetric variations, it seems to me that this is the most physic way to represent this type of structures.

Thanks again for your time, and I'll be glad to hear any other consideration.

Cheers,

Benjamín

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×