Hello all,

I've solved the problem. In the first approach (using a for loop to
calculate fcV), I had to multiply the computed fcV with the mass matrix,
following the below weak form:

[image: image.png]

However, we do not need such a multiplication, in the second approach as
the assembly of the entire matrix M*F is already done in the
compute_nl_term() function. Just wanted to share it in case someone else
can also make the same mistake.

Best,
Kübra


On Wed, Jun 16, 2021 at 8:30 PM Kubra Karayagiz (Alumni) <
[email protected]> wrote:

> Prof. Bangerth,
>
> Thank you for the prompt response.
>
> On Wed, Jun 16, 2021 at 7:33 PM Wolfgang Bangerth <[email protected]>
> wrote:
>
>> On 6/16/21 12:21 PM, Kubra Karayagiz (Alumni) wrote:
>> >
>> > Could you please let me know if my understanding of the usage of this
>> function
>> > is correct? (Note: the use of compute_nl_term() is necessary for the
>> > implicit-explicit implementation, due to the nonlinear functions)
>>
>> Kubra,
>> much debugging will be necessary to determine where the issue lies. I
>> don't
>> think any of us will be able to tell just by looking at the code.
>
>
> I totally understand that. I'll try to figure it out by myself.
>
> Have you
>> tried a simple case first, say where you start with c=constant and see
>> what
>> happens there?
>>
>
> I am sorry, I can't quite understand how it could be possible a case with
> c=constant, as it is the concentration variable that I am solving for. But,
> as I mentioned already my code works properly if I compute the function
> fcV(c) within a for loop.  (Based on comparisons with a benchmark study
> written in Prisms-PF software adopting deal.ii libraries).
>
>>
>> Separately, the problem you are trying to solve has fourth spatial
>> derivatives. How do you deal with this?
>>
>
> To prevent fourth-order, I split the Cahn-Hilliard equation into two
> (Equations (1) and (2) below). I first solve for mu (chemical potential),
> then use its solution while solving for c (concentration).
>
> [image: image.png]
>
> Thanks,
> Kubra
>
>
>> Best
>>   W.
>>
>>
>> --
>> ------------------------------------------------------------------------
>> Wolfgang Bangerth          email:                 [email protected]
>>                             www: http://www.math.colostate.edu/~bangerth/
>>
>> --
>> The deal.II project is located at http://www.dealii.org/
>> For mailing list/forum options, see
>> https://groups.google.com/d/forum/dealii?hl=en
>> ---
>> You received this message because you are subscribed to the Google Groups
>> "deal.II User Group" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to [email protected].
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/dealii/3391ca72-d1ca-dbb5-f35a-05185bd488aa%40colostate.edu
>> .
>>
>

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/dealii/CANnYN%2B2%2BEVWsxEG1Ok7TotvAm9wTV-m8yKYtEtSagwyPnejFHg%40mail.gmail.com.

Reply via email to