On 01/27/2014 09:02 AM, Nikolaus Rath wrote:
> On 01/27/2014 01:23 AM, Colin Cotter wrote:
>>
>>>> Sure. This is why it seems logical to me to constraint both trial and
>>>> test space by Laplace equation. Nevertheless I did not think it over a
>>>> much.
>>>
>>> For what it's worth, it seems logical to me as well... I just don't know
>>> how to impose the second constraint.
>>>
>>> So, if anyone could give me a hint or point me to a demo that shows how
>>> to constrain test functions, I'd be very happy.
>>
>> You don't need to constrain the test functions. This is actually the 
>> whole point of the Lagrange multiplier. You end up with an equation with 
>> the unconstrained test function, but if you choose a constrained test 
>> function, the Lagrange multiplier term vanishes.
> 
> I don't think that enforcing an arbitrary constraint on my solution will
> be equivalent to requiring the test functions to satisfy Laplace's
> equation :-).
> 
> 
> So I guess you are saying that if I require the solution to satisfy
> *Laplace's equation* via a Lagrange multiplier, this is equivalent to
> constraining the test functions to satisfy Laplace's equation?

... or does this require a different constraint on the solution?


Thanks,
Nikolaus

_______________________________________________
fenics mailing list
[email protected]
http://fenicsproject.org/mailman/listinfo/fenics

Reply via email to