convergence of Coco model

Discusses use of COCO, the process simulation and modelling software suite from AmsterCHEM, downloadable from http://www.cocosimulator.org

Moderator: jasper

convergence of Coco model

Postby Tom » 14 August 2014, 09:07

Hi,

I am not a very experienced user and have often Problems with convergences; so for example, the attached model is not converging. I suppose I have to Change some values in "Settings => flow sheet Options". What do recommend doing?
Thanks
Thomas
Attachments
Methanol aus allen CO2 dann DME - Kopie2.fsd
(501.07 KiB) Downloaded 725 times
Tom
 
Posts: 9
Joined: 28 July 2014, 15:55

Re: convergence of Coco model

Postby jasper » 14 August 2014, 10:34

(When in doubt: use Full Newton).

If you set the verbosity level from verbose to debug, you will get some more information, such as which is the variable with the highest residual between iterations. After a couple of iterations you will see that the recycle flow goes up in amount, quickly. Also the hydrogen flow on the recycle is pointed to as a poorly converging variable. This points in the direction that you are not venting or converging sufficient hydrogen.

You have a vent stream that is rich in hydrogen, but only 2% of the recycle. The reactor does not convert a lot of hydrogen either. So I would first check if the problem set up is realistic.

Another common technique is to break the recycle manually, e.g. before the mixer. Solving the problem then shows that you are recycling 729 kg/s on a feed rate of 890 kg/s, a lot of which is hydrogen. The conversion in your reactor is fixed, so your conversion will not increase with the hydrogen recycle. Again this point in the direction of being unable to vent or convert sufficient hydrogen.

Easily verified: let's increase the split factor for the vent, to say 100%. Even then you are not able to get rid of all the H2. Different approach: put a compound splitter in before the recycle mixer to get rid of all H2. Trying to convergence this shows that now you are having still a problem with build up of your recycle that consists mostly of CO2. You can also take this out using the compound splitter. Then it converges ok, and you are recycling just water.

Putting half of the CO2 back in (change CO2 split factor to 0.5) converges. 0.3 converges. 0.1 converges. Recycle flow rate goes up drastically.

You can figure out with parametric studies or manually when it stops converging. But the long and the short is: you are not purging enough H2 and CO2.
User avatar
jasper
 
Posts: 1128
Joined: 24 October 2012, 15:33
Location: Spain


Return to COCO (AmsterCHEM)

Who is online

Users browsing this forum: No registered users and 12 guests