DWSIM - Open Source Process Simulator / Discussion / Open Discussion: Calculate enthalpy for off-gas mixtures

stefan pofahl - 2013-12-13

Hi,

I'm interested in correct values for enthalpies of off-gas mixtures.
For this purpose I compare the simulation results of different software.
Up to now I and two others have calculated enthalpy values with the
following software:
- EES
-Cycle Tempo
- gPROMS
- DWsim
It would be nice, if others could also perform calculations with their
simulation tools. Or it would be interesting, if reference data of similar
off-gas mixtures is available to validate the DWsim calculation.

https://www.dropbox.com/s/1r11t0aq79jy6nd/enthalpy%20curves.png
The Excel-File that includes the data is available under the following address
(it includes also the DWsim calculations):
https://dl.dropboxusercontent.com/u/3639372/Comarison%20of%20different%20simulation%20software.zip

Regards,

Stefan

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Daniel Medeiros - 2013-12-13

Hi Stefan,

Are you using the latest build? The results from your sheet are not equal to the ones obtained through the graphical interface. I'm not able to use the add in here but I'll take a look at the code to see what's going on.

Daniel

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
- stefan pofahl - 2013-12-15
  
  Hi Daniel,
  
  I have not yet checked the results with the DWsim software.
  I hope that I find the time to perform the same calculation
  also with the GUI. I'm interested, if someone knows well established
  reference data for similar off-gas mixtures.
  
  Regards,
  
  Stefan
  
  P.S.:
  Any recommendations regarding the physical property model?
  
  If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Daniel Medeiros - 2013-12-18

Hi stefan,

I created a patch that fixes the enthalpy and entropy bug. Results are now similar to Cycle Tempo and EES.

To update your copy of DWSIM, just unpack this zip file replacing DWSIM.exe and DWSIM.pdb in your installation directory: http://sourceforge.net/projects/dwsim/files/DWSIM/DWSIM%203.0/Patch5099.zip/download

Regards,
Daniel

graph.jpg

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
- stefan pofahl - 2013-12-18
  
  Hi Daniel!
  That is great :-) Thanks a lot!!!
  My first impression was: Not much had changed, the differences are minor, if you do not change the property method PR. Then I restarted the computer and the results were still the same. In the next step I selected the method PR/LK and after this the result was equal to your plot. The strange thing comes now: I switched back to the original method PR, but now the result did not changed back to the old values. This was trigger for me and I started a comparison of the different models (the flash option is alsways the same "Nested Loops VLE). See attached graph.
  Regards,
  Stefan
  
  Last edit: stefan pofahl 2013-12-18
  
  comparison physical properties.jpg
  
  If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Daniel Medeiros - 2013-12-19

I'm glad you were able to get it working :-) I was trying to recalculate the sheet using F9 but it wasn't working, I had to recalculate everything manually.

As soon as you update your copy of DWSIM.exe and restart excel, it should load the new build, the results will remain the same only if the sheet isn't recalculated after that.

Daniel

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

stefan pofahl - 2013-12-20

Hi all,
there is a mistake in the excel file I shared with you in the initial entry in this thread.
The numbers, calculated by gPROMS, are correct. I just made a mistake: instead of H@35°C I took H@10°C and so on. I f you correct my mistake the gPROMS numbers are in accordance with the other calculation tools (see attached graph).
Regards,

Stefan

Comparison between Delta Enthalpy Calculations.jpg

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

stefan pofahl - 2014-01-29

Another update about the comparison, now also an AspenPlus result is included.

comparison of different software.jpg

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Daniel Medeiros - 2014-01-29

Thanks, stefan. I wonder where this difference comes from, if they're using an updated version of the PR EOS or if it is something else. The one implemented in DWSIM is the first version of the EOS, published in 1978. There have been some revisions after that.

Do you think that it is something worth taking a look at?

Last edit: Daniel Medeiros 2014-01-29

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
- stefan pofahl - 2014-01-29
  
  Hi Daniel,
  yes I think, it would be interesting to check the deviation between the initial PR EOS implementation and the current version. If you need to buy a textbook or some articles, I'm willing to contribute financially. Today I modeled the dew point with the PR EOS and compared the results in air with the "Arden Buck" formula and observed also an off-set. I will investigate further and will share my results here in the forum.
  Regards, Stefan
  
  If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
  - Daniel Medeiros - 2014-01-30
    
    Hi Stefan,
    
    Is it possible to separate the enthalpy calculation in ideal and non-ideal parts for the other models you've used to compare with DWSIM? This could help me a lot on tracking the systematic difference you've observed in your studies.
    
    Regards,
    Daniel
    
    If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

stefan pofahl - 2014-01-31

Hi Daniel,

unfortunately I do not have access on my own to any of the commercial software, I used for the comparison. I found some volunteers that provided these numbers.
But what I did today was: combining two EOS in DWsim to calculate the enthalpy difference. I computed the phase fraction with "Peng-Robinson-Stryjek-Vera 2 (PRSV2-VL)", because I observed that this EOS can predict the dew point of water in air quite accurate and afterwards I used the EOS "Peng-Robinson" (PR) to calculate the enthalpy (please have a look on my Excel file).

Regards,

Stefan

enthalpy of off-gas - dwsim based.xlsx

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

stefan pofahl - 2014-02-03

Hi Daniel,

I do not have direct access to the different commercial software, but I have an Excel file with some further detailed results from the gPROMS calculations (this file is added below).
This morning I combined the enthalpy calculation from DWsim and the phase fraction calculation from gPROMS and the result is: the DWsim enthalpy calculation is in line with gPROMS. Therefore the problems is probably caused by the erroneous phase fraction calculation. If you are interested in more numbers, I can offer two things: I ask my colleague, if he can calculate more / different numbers with the help of gPROMS or I can try to install the trial version of EES and do some calculation with EES.

I hope the attached files help already to go forward, kind regards,

Stefan

Last edit: stefan pofahl 2014-02-03

off-gas Mixtures.7z

phase fraction from gPROMs.jpg

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

stefan pofahl - 2014-02-03

It is also interesting to compare the vapor fraction for the four gas mixtures at 35°C, with different EOS. See the attached graph (best match with gPROMS is given for the EOS: Lee-Kesler-Plöcker. Stefan

vapor phase comparison.jpg

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Daniel Medeiros - 2014-02-03

Thanks stefan, the information about phase fraction is enough. I have a starting point now. :-)

Regards,
Daniel

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
- stefan pofahl - 2014-02-03
  
  With pleasure :-)
  
  If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Daniel Medeiros - 2014-02-04

Stefan,

the difference may lie in the interaction parameters used to calculate equilibrium. Can you list which ones are being used by gPROMS?

Thanks,
Daniel

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Daniel Medeiros - 2014-02-04

These are the ones being used by DWSIM. In the next release you'll be able to get their values from Excel too.

fig1.jpg

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

stefan pofahl - 2014-02-05

Hi Daniel,
at the moment my colleague is not available to look for the interaction parameters.
I will provide these numbers ASAP, but this may take a while.
Meanwhile I tried to calculate the vapor fraction with a simplified approach. I assume that the steam content at dew point is similar to air. If this is the case I can utilize the Arden Buck equation.
Regards, Stefan

phase fraction calculation.xlsx

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

stefan pofahl - 2014-02-06

Hi Daniel,
I received the interaction parameters for the AspenPlus calculations.
These parameters are the same for the PR and PRSK.
Regards,
Stefan

AspenPlus Interaction parameters.xlsx

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

stefan pofahl - 2014-02-06

I have tried to optimize the interaction parameters (IP),
could you indicate, if I understand the concept correctly.
It seems that the AspenPlus IP numbers do not change the results
significantly.
To optimize the IP I used Excel Solver Add-In.

INTERACTION PARAMETERS.xlsx

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Daniel Medeiros - 2014-02-06

Hi stefan,

Yes you've got the concept right but in this case it would be better to change the ones which are equal to zero. It wouldn't help much, though. It seems to be a limitation of the EOS itself, since my implementation is based on the original paper without any modifications.

Also, the interaction parameters are usually in the -0.1 - 0.1 range, they never get that high (in absolute terms).

Regards,
Daniel

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
- stefan pofahl - 2014-02-06
  
  Hi Daniel,
  
  thanks for the clarification. I was also not sure, if these huge numbers for the IP parameters make sense. The reason why I had not changed the other IP, was caused by the fact that these numbers are also zero in the Aspen Plus data base.
  I'm looking forward to the next implementation of the EOS "PR" in DWsim :-)
  
  Regards,
  Stefan
  
  P.S.:
  I noticed that there is a strange progression for the 10°C vapor fraction curve calculated by gPROMS. Maybe I ask my colleague to recalculate the values with a smaller step size for the Air-Lambda.
  
  If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

stefan pofahl - 2014-06-02

Hi Daniel,
thank you for the new Build 5265, now the phase fraction calculation ("PVFFlash2")
for EOS "PRSV2-VL" and "PRSV2-M" without proper initial value works fine and the result is matching the data calulated with the "Arden Buck" equation for humid air :-)
But it seems that the enthalphy calculation for the EOS "PRSV2-M" is broken.
Please have a look at the attached file.
Regards,
Stefan

Last edit: stefan pofahl 2014-06-02

dwsim forum example - enthalpy calculation.xlsx

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
- Daniel Medeiros - 2014-06-02
  
  Hi stefan,
  
  For these cases where the enthalpy is broken, try the IO flash algorithm. The NL result is not correct. Probably something during initialization, I'll take a look.
  
  Regards,
  Daniel
  
  If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Calculate enthalpy for off-gas mixtures

Simulate chemical processes using advanced thermodynamic models

Forums

Help

Calculate enthalpy for off-gas mixtures

Calculate enthalpy for off-gas mixtures

Simulate chemical processes using advanced thermodynamic models

Forums

Help

Calculate enthalpy for off-gas mixtures document.SUBSCRIPTION_OPTIONS = { "thing": "topic", "subscribed": false, "url": "subscribe", "icon": { "css": "fa fa-envelope-o" } };

Calculate enthalpy for off-gas mixtures