8 replies to this topic

[processengbd]

SIr,
I am a process engineer. Trying to simulate TEG dehydration unit in HYSYS.

My situation is as follows

In HYSYS example file (G-2) in dehydration unit stack column has been simulated as a modified distillation column which remove water from rich Teg to make lean Teg. In the column environment they used flash separator instead of column tray. I tried this method but was unable to converge the column.

I used an absorber and a heat exchager to to simulate the unit. It converged. And showed satisfactory results.

One of the previous post in cheresources.com an engineer used refluxed absorber to simulated the stack column.

So my question is what should be the actual method. Which simulation will provide the material and energy balance of the actual real world plant.

So my simulation
 Untitled.jpg   18.44KB   75 downloads

An engineer from cheresources.com
 22-12-2009 11-36-43.png   135.26KB   103 downloads

Hysys example file
 Hysys G -2.jpg   48.17KB   89 downloads


 Hysys G -2 Column Environment.jpg   29.54KB   58 downloads

Thank you
Processengbd

Edited by processengbd, 08 November 2012 - 07:59 AM.

[processengbd]

Today used the HYSYS example file procedure to simulate the unit. The column converged. It showed satisfactory separation.

I took a custom column and added equipment according to example file. Then added two specification one reboiler temperature and condenser temperature.

The man reason I recon of using a flash separator in the column is that. Glycol and water has vastly different volatility simple flashing at elevated temperature would remove water from glycol. Stack column is basically used to recover some Glycol from the vapor. And stripping gas is used to get purer glycol regeneration. So in simulation we do not need to simulate stack column and stripping gas.

Thank You
Processengbd

[mattb]

nowadays almost all simulators (Winsim, Promax, ProII, HYSYS...) allow to simulate a dehydration unit,
take care to select a proper thermodynamic model for VLE and enthalpy,
differently you may get as solution some unphysical values...

[Ceng]

Processengbd

TEG regenerator normally required very few theoretical stages (2 is typical), and also a very low reflux is needed.
So I guess different modeling you stated is because of what I said above. In one case the designer hasn't considered any reflux and in another case he considered the tower with 3 theoretical stages as a flash drum with reboiler and condenser.
Also note that regeneration of the glycol by atmospheric distillation at about 200 °C achieves about 1 weight % water in regenerated glycol.
To achieve less than about 1 weight % water in TEG, there are several special regeneration processes available. Examples are use of stripping gas (which has been shown in one of your examples) with or without an extra column, Cold finger , Drizo and the Read Cycle .GPSA and Kohl are general references for dehydration processes.

[processengbd]

Thanks everyone
Mr Mattb I used Glycol Package for my simulation. HYSYS example file uses Pengrobinson. I think Glycol package will give me best possible result.

[processengbd]

But I have a question in Dehydration unit simulation several component other then Glycol has been used. So for natural gas and water should I use separate property package?

[Ceng]

Glycol package for dehydration units with glycol will give you the best results.
But if you have other Units such as sweetening, compression, condensate stabilization etc. you should change your property package for reasonable results (for that units).
For example Amine Pkg for sweetening is suitable.

[momen1986]

i preparing to gas dehydration cycles and iam urgently needed hysys simulation for gas dehydration and factors affecting

[Ceng]

Please start a new topic and be more specific