Results 1 to 4 of 4

Math Help - Convergence of the sequence of discretized solutions of a variational problem

  1. #1
    Newbie
    Joined
    Jun 2009
    Posts
    3

    Convergence of the sequence of discretized solutions of a variational problem

    Consider the Galerkin discretization of an abstract variational problem where the Hilbert space V is separable.
    http://en.wikipedia.org/wiki/Galerki...stract_problem

    Each of the subspaces Vn is generated by the first n terms of a sequence of elements of the separable Hilbert space V. This sequence is such that each of these subspaces will be dense in V. The problem is proving that there is a subsequence of the bounded sequence of discretized solutions , that converges weakly to the solution of the variational abstract problem , then proving that the same subsequence converges strongly and finally proving that the sequence of discretized solutions converges to the solution of the variational abstract problem .
    Follow Math Help Forum on Facebook and Google+

  2. #2
    Super Member Rebesques's Avatar
    Joined
    Jul 2005
    From
    At my house.
    Posts
    538
    Thanks
    11
    Let's see... Let (u^{(n)}) be the Galerkin approximation. Identify elements in each subspace V_n with their natural injections into V. Coercivity of a implies that c||u^{(n)}||^2\leq a(u^{(n)},u^{(n)})=f(u^{(n)})\leq ||f||||u^{(n)}||, which gives that (u^{(n)}) is bounded in V and must therefore possess a subsequence, denoted by (u^{(n)}) also, to weakly converge to u\in V. Since a(\cdot,v) is linear, we have a(u^{(n)},v)\rightarrow a(u,v) which means u is also a solution of the initial problem. Now, to demonstrate the strong convergence of u^{(n)}, notice that coercivity and Galerkin orthogonality imply c||u^{(n)}-u||^2\leq a(u-u^{(n)},u-u^{(n)})=f(u-u^{(n)})=0(n).
    Last edited by Rebesques; July 17th 2009 at 12:29 AM.
    Follow Math Help Forum on Facebook and Google+

  3. #3
    Newbie
    Joined
    Jun 2009
    Posts
    3
    Thank you for the answer. I havenīt been able to prove as yet that, as you state in your answer, for any linear form and for any weakly convergent to u, we will always have that . Can this be infered from the Lax Milgram lemma?
    Follow Math Help Forum on Facebook and Google+

  4. #4
    Super Member Rebesques's Avatar
    Joined
    Jul 2005
    From
    At my house.
    Posts
    538
    Thanks
    11
    Not really. Just the definition of weak convergence.
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. Replies: 4
    Last Post: October 3rd 2010, 04:54 AM
  2. Variational formulation problem
    Posted in the Advanced Applied Math Forum
    Replies: 1
    Last Post: June 16th 2010, 01:15 PM
  3. Replies: 6
    Last Post: October 1st 2009, 09:10 AM
  4. Uniqeness of weak solution to variational problem
    Posted in the Differential Geometry Forum
    Replies: 1
    Last Post: March 18th 2009, 10:06 AM
  5. Replies: 4
    Last Post: October 17th 2006, 11:54 AM

Search Tags


/mathhelpforum @mathhelpforum