DWSIM Development Blog - Comments

order essays: ...

nospam@example.com (order essays) — Wed, 08 Sep 2010 23:15:35 -0300

Correctly written custom writing would bring you a chance to reach a success. Nevertheless, customized term papers accomplishing would use all free time. Thus this would be executable to order essays to foreclose it.

best online papers: ...

nospam@example.com (best online papers) — Wed, 08 Sep 2010 23:15:11 -0300

Even if you strive a lot, it will not be a guarantee that you will gain a success. Sometimes, students who make nothing have higher grades. That occurs because they buy custom written research papers. Probably you have to do the same?

ElviraMoore27: ...

nospam@example.com (ElviraMoore27) — Sun, 05 Sep 2010 19:24:49 -0300

I opine that to receive the loans from creditors you ought to have a firm motivation. However, once I've received a car loan, just because I wanted to buy a car.

anon: DWSIM 1.7 Build 3868 released

nospam@example.com (anon) — Mon, 09 Aug 2010 07:21:41 -0300

I tried the new source code on VS 2008 pro, SP1, I get the following error, 'the tools version '4' is not recognised', what is the problem?

Daniel: Plugin system for Utilities

nospam@example.com (Daniel) — Wed, 28 Jul 2010 09:53:54 -0300

Hi KP,

I sent you an e-mail.

KP: Plugin system for Utilities

nospam@example.com (KP) — Wed, 28 Jul 2010 00:56:25 -0300

Hi Daniel:

I did a little testing with the phase envelope utility with an equimolar mixture of C1, C2, C3, N2 and O2 and the PR EOS. DWSIM seems to predict the dew point curve up to 5 bars of the critical pressure. The bubble point curve and critical points are not calculated at all. Can you replicate this?

KP

Daniel: Plugin system for Utilities

nospam@example.com (Daniel) — Thu, 22 Jul 2010 09:49:22 -0300

Hi KP!

Yes you can use C# and VB.NET directly in Silverlight, but what I don't know is if it supports all .NET namespaces, mainly Generics... I'll take a look.

About the column, I setup that problem before changing the algorithm and didn't dig in into the code to see what is happening but I can confirm your observations. Check the specifications, if you saved the simulation before I did the modification, they will be inverted - you have to setup them again.

Kind Regards,
Daniel

KP: Plugin system for Utilities

nospam@example.com (KP) — Thu, 22 Jul 2010 01:22:40 -0300

Hi Daniel:

Lots of progress! Keep up the good work. I had an idea that might be worth some consideration. For the next version, what do you think about implementing the whole application in Silverlight? To my knowledge, you can use C# directly in a Silverlight app ... it might be a easy way to deploy and port dwsim to multiple plaforms (win and mac for sure) and linux (moonlight).

Also, I've been testing out dwsim a little ... I don't know if there is a problem with petroleum_dist example file. Minor changes to feed oil temperature (+ 5 K) will never converge the column solver. I thought maybe it was slow, but I waited for 5+ minutes ... is this expected behavior? And also I'm surprised that the column does not converge with the IO method. I haven't dived through the source yet, but maybe you have some insight?

Regards,
KP

Balaji: Important information about the new IO Flash in DWSIM 1.7

nospam@example.com (Balaji) — Sun, 11 Jul 2010 21:57:08 -0300

Hai Guys Daniel and KP,

You guys are rocking. Really setting a great example of how world should function.

Thanks for the Great work.

Daniel: Important information about the new IO Flash in DWSIM 1.7

nospam@example.com (Daniel) — Sat, 10 Jul 2010 09:59:49 -0300

Yes KP, I did what they suggested bu the algorithm gets a little lost in the R convergence loop when it finds a single phase solution and takes longer to converge.

It just needs some fine tuning, the rest is working well.

thanks again!

KP: Important information about the new IO Flash in DWSIM 1.7

nospam@example.com (KP) — Sat, 10 Jul 2010 02:29:14 -0300

Hi Daniel:

A couple more suggestions for the IO algorithm. If you're doing any broyden updates ... delay them for atleast 3-4 iterations. The acceleration works much better in this case.

The single phase problem is solveable. Lookup this paper "Efficient flash calculations for chemical process design — extension of the Boston–Britt “Inside–out” flash algorithm to extreme conditions and new flash types" - by Parekh and Mathias. You can also try introducing the pseudo-roots ... that the algorithm will converge most of the time except you'll get negative vapor fractions (or vapor fractions greater than one), indicating that you need to perform the calculation in the one phase region.

Hope this helps!
KP

Daniel: Phase Envelopes with new flash schemes

nospam@example.com (Daniel) — Tue, 06 Jul 2010 17:12:56 -0300

Hey KP, thanks for the tip, I found the paper!

Regarding the problem with stability, I just discovered a bug in the initial temperature estimate for the dew point curve. It is fixed now. I just have to solve the trival root problem, currently I'm returning ideal K-values when that happens.

KP: Phase Envelopes with new flash schemes

nospam@example.com (KP) — Tue, 06 Jul 2010 16:28:59 -0300

Hi Dan:

Phase envelopes look good! There are a couple things you might want to consider using in conjuction with the inside out method. The inside-out method works better with a method of predicting pseudo-roots with an EOS. This is should help with the trivial-K value problem. "Effective utilization of equations of state for thermodynamic properties in process simulation" by Mathias et al. is a simple approach. Let me know if you can't find this paper.

Another thing to consider is clipping the T or P excursions in the outer loop during the early iterations. I've found that clipping the changes to 20-40% helps the algorithm a lot. You might end up with a few more iterations on the easy cases but you're more likely to converge on the hard ones.

And yes, if all else fails, you do have a nested T-P algorithm as well! Good luck!

Martin: Inside-Out Flash in DWSIM

nospam@example.com (Martin) — Fri, 25 Jun 2010 17:52:46 -0300

Nice work, hopefully i can test this on my ethylene sim. I tried simulating a 250 bar system, but the phase envelope didnt look all that good from 0 - 250 bar.

Daniel: Comments about unreliable thermodynamics on DWSIM...

nospam@example.com (Daniel) — Thu, 20 May 2010 17:23:50 -0300

Of course, Marco. Don`t worry, I'll work on that too.