Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Writing Finite Element Code for Structural Analysis

  1. May 21, 2012 #1
    Structural Analysis-Writing Finite Element Code,

    Dear all,

    I have written a code ( in fact it is a software) for 3D finite element structural analysis. While developing the code I found that assembling global stiffness matrix is quite complicated. The complication is even more when we want to reduce the degree of freedom of some of the nodes. Also for giving boundary condition, when we set one or two components of deformation rather that all the three. Now these parts of my code has become so complicated even to myself.
    My code works well now but I am wondering if the complexity is usual or if I am doing the job in an ineffective way. Since I have not discussed my methods with anyone else, I thought it would be helpful if I post the problem here. Perhaps some of you have experiences to share.

    Regards,

    Hassan
     
  2. jcsd
  3. Apr 6, 2014 #2

    mdn

    User Avatar

    Dear Hassan, I also developed finite element Solver and automatic mesh generation program in Java, but for 2D only. I getting accurate result for elliptical, parabolic and hyperbolic governing equations. Your description is short.
    Please describe complete problem.
    Could you give information of following questions?
    1. What type of boundary condition you applied: Dirichlet or Neumann?
    2. which technique, you are using for elimination of boundary nodes?
    3. which language, you used for software?
     
  4. Apr 6, 2014 #3
    Thanks mdn for the reply.

    The post is almost two years old and the problems are solved now. For your information the answer to your questions are the following:

    1. I use a variety of boundary conditions. In fact they are not limited to the boundary. I may have Dirichlet/ Neumann/Periodic boundary conditions, or "sliders". By sliders, I mean a certain component of the displacement at a defined region is fixed ( usually to zero).

    2. I assemble the matrix for the unknowns only. this may make the matrix construction a bit complicated though but I am satisfied with it.

    3. I use Java programming language. The language may not be advantageous for the solvers but since I have also developed the visualization software and that is easier in Java using Java3D library, I keep using Java for the solver too.

    Thanks again.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Writing Finite Element Code for Structural Analysis
Loading...