[Getdp] ... Formulation question ...

mkoch at gvtc.com mkoch at gvtc.com
Fri Mar 2 00:02:14 CET 2007


Hello Ruth,

thanks to you, too, for replying so quickly. I just tried your  
suggestion, but it does not lead to success either.

So, Galerkin{[Spd[]*Norm[Dof{d Phi}],{Phi}];...} fails with:

"GetDP: Error: Wrong type of argument for function 'Norm'"

The same applies to Galerkin{[Spd[]*Norm[Dof{Grad Phi}],{Phi}];...},  
by the way.

And Galerkin{[Spd[]*Dof{Norm[d Phi]},{Phi}];...} fails with:

"GetDP: Error ('E:/Sapphire  
Systems/GMsh-GetDP/LevelSet/LevelSet-2D.sol' line 66): parse error on  
'['"

The same applies to Galerkin{[Spd[]*Dof{Norm[Grad Phi]},{Phi}];...},  
by the way.

As was mentioned earlier, I really am beginning to think this problem  
has to do with some sort of need for linearity in Galerkin{}, and  
Norm[] is simply not a linear function.

Regards,

Matt Koch



----- Message from r.sabariego at ulg.ac.be ---------
     Date: Thu, 01 Mar 2007 22:28:43 +0100
     From: "Ruth V. Sabariego" <r.sabariego at ulg.ac.be>
Reply-To: "Ruth V. Sabariego" <r.sabariego at ulg.ac.be>
  Subject: Re: [Getdp] ... Formulation question ...
       To: mkoch at gvtc.com


> Hi Matt,
> You are right, you need to have Dof{}, otherwise GetDP thinks Phi is known.
> Have you tried this?:
>
> Galerkin{ [ Spd[]*   Norm[Dof{d Phi}]  , {Phi} ] ... }
>
> Regards,
> Ruth
>
> mkoch at gvtc.com wrote:
>> Hello Olivier,
>>
>> thanks for the follow-up. Yes, I did see your comment about the following:
>>
>> Galerkin{ [ Spd[]*    Norm[{d Phi}]  , {Phi} ] ... }
>>
>> I guess it is only different from what I was trying in the Dof{} term, i.e.:
>>
>> Galerkin{ [ Spd[]*Dof{Norm[{d Phi}]} , {Phi} ] ... }
>>
>> However, if I leave out the Dof{} term, won't that mean that Phi is  
>>  no longer treated as a variable to be solved for, but rather as   
>> something  given already? In other words, I won't be solving the   
>> right problem?
>>
>> Thanks,
>>
>> Matt Koch
>>
>>
>> ----- Message from castany at quatramaran.ens.fr ---------
>>     Date: Thu, 1 Mar 2007 20:19:16 +0100
>>     From: Olivier Castany <castany at quatramaran.ens.fr>
>> Reply-To: Olivier Castany <castany at quatramaran.ens.fr>
>>  Subject: Re: [Getdp] ... Formulation question ...
>>       To: mkoch at gvtc.com
>>
>>
>>
>>> On Wed, Feb 28, 2007 at 04:24:58PM -0600, mkoch at gvtc.com wrote:
>>>
>>>> Hello Olivier,
>>>>
>>>> and thanks for the quick response. I was afraid you were going to say
>>>> something like you did. Is there at least a way to define Norm(Grad
>>>> Phi) as a quantity somewhere, say call it NrmGrdPhi, (perhaps in
>>>> Quantity under Formulation), and then use it in the Galerkin term. So
>>>> something akin to this:
>>>>
>>>> Formulation{
>>>>   Quantity{NrmGrdPhi; [Norm[Grad Phi]];}
>>>>   Galerkin{[Spd[]*Dof{NrmGrdPhi},{Phi}];In Vlm;Integration Int;}
>>>> }
>>>>
>>>> I suppose this would be cheating, too, but I got to give it a try. Any
>>>> thoughts?
>>>>
>>> Hello,
>>>
>>> I went on the GetDP mailing list this evening and I don't see the answer
>>> I posted in response to your message yesterday evening...  maybe I did
>>> something wrong.
>>>
>>> So :
>>>
>>> did you get my answer where I was saying that you could use a non
>>> linear solving scheme with : Galerkin{ [ Spd[]*Norm[{d Phi}] , {Phi} ]
>>> ... } ?
>>>
>>> Regards,
>>>
>>> O.C.
>>>
>>>
>>>
>>
>>
>> ----- End message from castany at quatramaran.ens.fr -----
>>
>>
>>
>>
>>
>> _______________________________________________
>> getdp mailing list
>> getdp at geuz.org
>> http://www.geuz.org/mailman/listinfo/getdp
>>
>>
>>
>
>
> -- 
> Dr. Ir. Ruth V. Sabariego
> University of Liege
> Dept. of Electrical Engineering (ELAP), Institut Montefiore
> Sart Tilman Campus, Grande Traverse, 10 (B28), B-4000 LIEGE, Belgium
> phone: +32-4-3663737 -- fax: +32-4-3662910 --
> http://elap.montefiore.ulg.ac.be/


----- End message from r.sabariego at ulg.ac.be -----