Professionally I work at BPT and focus on creating value adding App for Process Simulation. You can find more info on those on the BPT website. Below are posts that should help HYSYS, PetroSIM and UNISIM users alike in their day to day challenges to produce accurate yet fast models efficiently.

Release the true potential of your staff Apps for Process Simulation

Sunday, 24 May 2015

Did you know series: How to properly use the Gas Cleaning Package

This week I was reminded that maybe not everybody realises how much complexity is behind the gas cleaning package and what the consequences are. Let me jump to the consequences first, in case you don't read this post to the end:
 - There is a limited set of solvents and solvent mixtures available, if the mixture you need is not in the list, don't use the gas cleaning package. There are two ways round this. The first is to ask Aspentech if they happen to have a property package for your system, just might be lucky. The second is to use Aspen Plus and starting from the package that is closest to what you need, add the missing component(s). This should work reasonably well if you have a mixture for which there are existing packages to the individual components. You will be missing some interactions, but the result should still be reasonable. We (BPT) would be happy to do this for you (for a fee of course).
- You MUST specify the column geometry. A lot depends on the liquid and vapour residence time and the interfacial area, you only get correct values with a correct geometry.

The gas cleaning package relies on three complex phenomena. At the hart is an electrolyte thermodynamics package that accounts for "normal" vapour / liquid equilibrium, but also for dissociation of various molecules into ions. The second part is that some of the reactions that take place are not infinitely fast, so you have to account for reaction kinetics. The third factor is the mass transfer where the two film theory is used to calculate the transfer rates from one phase to the other. In some cases it get's really complex because the reaction rates are really fast, but not fast enough to consider equilibrium. In those cases, the film in which the reactions occur must be split up into multiple slices.

There are elements in the user interface of the column that give you and idea of the complexity and allow you to do some tuning if needed. After having installed a gas cleaning fluid package you can install the absorber column. When all the basic configuration is done, go to the Parameters tab / Acid Gas page. Let's stick to the efficiency model, the other model goes even more complex. On this page you specify the geometry of the column.

The picture shows a valve tray geometry, but you can change the "Column Type" to Packed or other internal types. On this page you enter the actual geometrical data. Once that is done, converge your column and check if you get the expected result. If you are in the neighbourhood, you can move on to tuning things. If the result is really far off, you probably should check you geometry data or amine composition or so.

For tuning you click on the "Advanced" button. You'll be presented with the following view:

The coefficient you are most likely to want to change is the "Interfacial Area Factor". You would only touch the other tuning Parameters if you have a thorough understanding of why one of those coefficients needs to change. At the bottom you can select the reaction set to be used. I is highly unlikely you'd want to change that. You can hit the view button to get an idea which reaction are taken into account.

This should give you some idea of what is behind the gas cleaning package. Hopefully enough to heed my advice at the start!

One last thing, in the end, the solution algorithm used as a whole for the column is still one of the classical HYSYS solution algorithms. Two tips if your column refuses to converge:
1 - If the equilibrium Errors do not nicely go down to a low number, try modifying the damping factor. A value of 0.2 (instead of 1) should help, in some cases you may want to go as low as 0.05 or so.
2 - If the mass and spec error do not converge, you may want to try the "HYSIM Modified Inside-out" algorithm. For the absorber there aren't any specifications to set, so there shouldn't be any specification issues. The one thing to possibly check is the pressure, if it is way off from the correct value, that might also an issue.

No comments:

Contents By Category