Differences from PVT report and HYSYS simulation result

[Yeonchan Jung]

Hi, all,

I am a newbie in the HYSYS software and desperately in need of a few adcvices.

I am simuating a onshore production facility in Kazakhstan using the HYSYS SW and now a little confused because there is a big difference in the production oil densities which I got from PVT report and simulateion result.
- Degassed oil density from PVT report (20^oC, 1 bara) : 849 kg/m³
- Liquid phase density from HYSYS simulation (20^oc, 1 bara) : 728 kg/m³
- Molecular weight in PVT report : 215 / Molecular weight in HYSYS : 138.4

I tried other PVT reports from other production field to compare with this error.
The result was around 5% lower densities in HYSYS than in the PVT reports, but not that much as above.


I doubt that there was a mistake while translating the PVT report in Kazakstan language into English.
It seems to me that there are too much light components in the PVT report to make 849 kg/m³ density.
- C6 : 7.4% / C7 - 15.5% / C8 - 16.7% / C9~C15 - 34% / C16~C20 - 10% / Above C20 - less than 1% each

Am I understanding wrong about the HYSYS and PVT report or are there any points that I missing?

One more question, if the PVT report is wrong, is there only one way to correct the mistake, which is to have another PVT analysis? Or, Is there a good way that I can simulate the 849 kg/m³ density and molecular weight of 215, using the HYSYS simulation without the PVT report?

Any piece of advice will be highly welcomed.
Have a good day and happy new year.

YC Jung.

 

[rotw]

PVT stands for?

Life is not about waiting for the storm to pass. It's about learning dance in the rain.

 

[georgeverghese]

Check that the PVT analysis report has been signed off by someone authorised with audit responsibility in that organisation. Large errors may be expected if the fluid P/T are around the critical region. On the other hand, the errors may also be due to incorrrect characterisation of some of the psuedo components.

 

[MortenA]

liquid density is often the most difficult property for a EOS to get right. Try changing the method for calculating the density. Default would be COSTALD - an emperical method overriding the density calculated by the eos. The EOS result would usually be more off than COSTALD. You can try to switch to the EOS with the Raclet option - but this wil require some additional info. You have to have this infor to see improvement, Dont think that the "estimaiton tool" will improve - because it uses the COSTALD method to etimate the coefficient. Note that the HYSYS metod is SIMILAR to the PVTsim (CALSEP) called Pennelux but they use another factor so the values differ by a factor -1000 (its been a while but i think its the HYSYS values that equals the PVTsim values/-1000 - but you have to check this yourself.

You can also examine the pvt report to see if the experiments were conducted in a manner so that the pressure were let down in more than one step. Then you should do your simulation the same way.

To answer another of your questions: Most likely its HYSYS thats off - but it may be a point in between (the measured value, the characterization, the composition etc.

You may see if you can find a charaterization that tells you the upper/lower bounds for each component. This could lead you to find a composition that gives a better match - but it will give you lots of questions so you will have to know what you are doing

Best regards, Morten

 

[MortenA]

otwr: Pressure, Volume, Temperature

 

[rotw]

Thanks MortenA.

Something I do not get.

Molecular weight in PVT report : 215 / Molecular weight in HYSYS : 138.4

How come? Mol. Weight is independent from EOS.
Thus you are not comparing same gas fluid mixtures.

 

[MortenA]

Hmmmm i didnt notice that and it sound strange. Are OP sure that the hypo components in HYSYS are defined correctly and that the concentrations are entered correctly?