No replies to this topic

[shvet]

Hi, forum

Sorry for a lot of text, but I've tried to be as clear as possible.
I'm newbie in Aspen Plus simulation and it would be great if anybody more experienced can help me or redirect me to another forum.

I believe in golden rule of chemic technology "garbage in => garbage out".Before I had worked with Pro/II and Aspen Plus and actually (as I thought/supposed) understand how it works but now I got stack. Unfortuantly I can't ask Aspen's support or to be more exact I've tried to ask other questions but I've not got any response. There's only retail office in my country and my company bought only 1 workplace (~50k$/y). We are too small to be worth spent time.

I'm working on rather simple distillation simulation - naphtha splitter. Straight run naphtha and hydrotreated naphtha mix together and C5-fraction goes through column reflux drum. It's existing equipment and I have to find out what the max capacity is. I have no ASTM D86 curves but I have individual components composition analyses from C3 to C12 with rather high accuracy. Undefined components is no more 1% weight total. All sounds good but I can't start to make simulation because can't complete initial inputs.

I made list of components and chose thermodynamic method (Aspen-SRK). But I can't start to draw flow diagram because of this kind of error messages:

"STRUCTURE FOR COMPONENT [component name] HAS NOT BEEN DEFINED.
 PCES CANNOT USE GROUP-CONTRIBUTION METHODS TO ESTIMATE MISSING PROPERTIES
 USE THE STRUCTURES PARAGRAPH TO DEFINE STRUCTURES OF THIS COMPONENT."

For all kind of components! Including H2O, propane, butane etc.

"WARNING IN PHYSICAL PROPERTY SYSTEM
PARAMETER PLXANT (DATA SET 1) FOR COMPONENT [component name]:
UNREASONABLE VAPOR PRESSURE CALCULATED AT NORMAL BOILING POINT.
VALUE SHOULD BE WITHIN 10% OF 1D5 N/SQM.  PLEASE CHECK YOUR INPUT
FOR THE ANTOINE PARAMETERS."

For only some of high-boiling components.

As I understood there is two types of databank in Aspen Plus:
- pure-component databank (for example AP-PURE or AP-HYSYS)
- binary parameters databank (for example Aspen-SRK or HYSYS-SRK)

And there are binary interaction parameters in binary databank (DB) only for simple hydrocarbons (HC) - propane, butane etc. There are no binary pairs for high-boiling hydrocarbons in databanks. As I understood software engine shall take binary interaction parameters from chosen binary databank (for example Aspen-SRK) and if there is no binary coefficient for some binary pairs then estimate them with algorithm using pure-component's properties from pure-component's databank. If later I will find results of such model are too deficient I can change thermodynamic model (for example from Aspen-SRK to HYSYS-SRK) or even try to estimate binary coefficients for missing pairs with universal model like UNIFAC/UNIQUAC.

I fill that I don't take into account some simple detail, something obvious. Please correct me or confirm:

1. I can add to one simulation different components from different pure-component DBs. For example for high-boiling HC if I've found one component in AP-PURE DB and other component in AP-HYSYS DB I can add both of them in one simulation's components list. I will not face any problem later.

2. I have to draw molecule structure for all components using molecule editor for all components. For all?! Including most simple H2O, propane etc., all 150 components?! Are you kidding me, it's so much time.

3. I have to exclude from simulation components with invalid Antoine parameters.

4. I should change default parameters in "Estimation" window and change estimation methods. For high-boiling HC? For all HC?

5. I should input missing UNIFAC constants Q and R for high-boilng HC. Only if I'm planning to use UNIFAC estimation for missing pairs? In any case?

 

Simulation file

Edited by shvet, 13 May 2016 - 11:54 PM.