Jump to content
GMS, SMS, and WMS User Forum

karunakar

Members
  • Posts

    16
  • Joined

  • Last visited

  • Days Won

    1

Posts posted by karunakar

  1. hello every one,

    i have been doing the following model for some time.

    the basis of the model is study of impact by artificial reefs on wave induced local currents

    we have designed a model oriented for peak wave direction

    We have a model of 5m resolution suggested by the tool box for the wave periods and runup.

    the wave maker is kept at a depth of 10m.

    The cell string on sides are purely absorbing and land has a value of 0.1(as suggested in the tutorial).

    mostly it is in accordance with the bouss2d tutorial.

    it has given some decent results, till breaking zone, post which does not agree so good with the site data.

    A model has been constructed with grid size of 1.5m to depict reef width of 3m clearly, with some smoothening in bathymetry.

    this doesnot show up any result.

    i've checked the timing of wave maker, model control to we have increased the simulation time upto 3000 steps. but no result has been updated post simulation. what could be the problem

    Looking for some light.

    Thanks in advance

  2. hi there.,

    is it possible to sum two WSE data sets spread in time (between say some time bounds) on a sampled time interval.

    though resampling option is there is sms, i'm concerned with merging two different data sets.

    can any one suggest some work out.

    Thanks in advance,

    Karunakar kintada

  3. hello there,

    of recent i've encountered a typical problem.

    i've built a mesh for a domain and done a lot of manual editing on it.

    unknowingly i've lost the land boundary node string, i've tried changing colors, but of no use.

    ocean nodestring is visible in blue. so 'm sure land node string is gone.

    is there any automated way to get it back (please don't suggest i've to manually do it).

    Thanks in advance,

    great day.

    --karunakar kintada

    Scientist

  4. hi there.,

    'm sorry, most of my questions seem weird.,

    but i would like to ask.,

    is there any way can compile the padcirc on my computer.

    i've got good hardware.

    my WorkStation is 8 intel xeon core based and 'm currently running xp.

    my colleagues also have the same config - 3 more machines.

    i've got padcirc code from ftp (one has to place a request for the same).

    i've read through make complrflags file, but unable to locate windows xp environments.

    now that i'm interested in compiling the same on this machine and use mpich2 and accelerate the processing.

    can some one sort out this issue.

    thanks in advance

    • Like 1
  5. thanks Zundel,

    the select option was not working earlier and i updated sms to 10.1.11 and found the things back to normal again.

    for info,i was unable to open the fort.14 file was also(CRC error).

    with xyz values of the previous mesh, i now got the model back.

    thank you all.

    You mention that you have positional data. Does this include all the nodes from the fort.14 or is it only partial. Do you have the nodestrings or the domain extents in a map file?

    If you have all the points, simply triangulate the points and then delete the triangles that are outside of the domain. If you have a domain definition as a map file (or shape file), you can select all the triangles outside of the domain and delete them in a single step.

  6. thanks aquadraht.,

    i dont have the any of the files working because of a problem with my os - encryption.

    sometime back i've converted the mesh file to scatter point set and saved it as a pts file(saved in my mail).

    the earlier mesh provided me good validation, so i want to continue with the same.

    but now i'm stuck up with this problem.

    is there any way that i can eliminate triangulation in the polygons that were attributed land??

    Hi karu

    You can try to just load the point and triangle informnation into SMS. Open the 2dm file in text editor and delete everything from "BEGPARAMDEF" to the end of file and save as new *.2dm. All important information (Node, Tringles, Nodestrings and Material type) should still be stored in the file. Now try to reread the file into SMS.

    ciusi

    D

  7. hi rusty.,

    sry for long break.

    any suggestions??

    Hi Karunakar,

    You bring up a good question with LTEA using UTM. I don't have a good answer and am asking around to see what I can come up with.

    On the other items...

    I am not sure how long it would take to mesh something similar to the WNAT. If you are talking about the 2.2 M node mesh, it may take a significant amount of time. I would recommend using the 64-bit version of SMS for something this large and obtain as much RAM as possible (I run with 12 GB and others here have over 20).

    SMS goes into unresponsive mode when it is busy doing calculations or other things. It doesn't necessarily mean that it is stuck permanently but that it is busy. I would let it run for a couple hours on domains less the 1M nodes and overnight for domains larger than this. If it doesn't become responsive by this time, please contact support@aquaveo.com and send them your files so we can take a look at what is happening.

    Sorry, SMS doesn't currently have an autosave feature.

    Thanks for questions. Hopefully, we can come up with a good answer on your main question.

  8. hi there.,

    i have nodes positional(xyz) data from a previous model(currently project file corrupted).,

    is there any chance i can regenerate the mesh.

    the point here is, if i try importing the file as mesh, the triangles are generated even on polygons with land attributes(eg main land and islands).

    any suggetions please.

  9. hi all.,

    i've been thinking of this for the last couple of months.

    Some time ago, i was referring to a paper by Westerlink, parrish and others regarding the mesh generation using ltea + cd for WNAT model.

    and in SMS the LTEA asks for the projection to be in UTM coordinates (ref Meters not in degrees).

    sometimes it is very difficult to convert a bigger domain into said coordinates, where it may span into different utm zones.

    even if we manage it out with Transverse mercator some how, after meshing conversion back will distort elements too much at borders, particularly towards corners of the domain selected.

    any suggestions on this topic are most welcome.

    for Rusty:

    hi jones, i would like to ask you regarding how much time it would take to mesh a domain similar to that of WNAT.

    most of the times the SMS package goes "Not Responding" and no feed back from "paving interior(X)" in the bottom left of the window.

    and the toolbars get distorted after mesh generation, sometimes which if adjusted make the software unstable.

    is there any auto save sort of option available with SMS.,

    excuse me if i'm ignorant of any options and rules.,

    thank you all.

×
×
  • Create New...