Tuning EOS parameters

[hoanglongbk88]

Shakepear, Do you have a spreadsheet to calculate the dewpoint pressure based on EOS? I am trying to write on excel to calculate DEwpoint pressure base on Peng-Robinson By visual basic application (VBA). but the algorithm of calculating the DEWPOINT is not clear as flash calculation ( vapor-liquid) that is written sucessful by EXcel.

[Shakespear]

I would do the following. Take my best results and go to my dissertation adviser and say.

1) "I did my best to tune the data within the time I had"
2) "Here are the problems I see"
3) "Here is how they would effect my results"

In other words, you need to show that you understand the effect of the error on your simulation. Thus you don't just dump data in, get results and say I am done. In the future you will know what to do, send them to a PVT expert.

With your time constraint I would let it go. Trying to learn a PVT program is not easy as each has its own problem. I just found out some limitations that post 2007 version of PVTi has and anyone not knowing this will have lots of fun trying to get this particular option to give good results. I will not be specific, but if you have a legal copy JUST CALL Schlumberger for information about different version problems.

Regarding calculating the Dew Point, again, that one reference I gave before had a lot of good papers on Condensate Reservoirs and has what you need. But are you sure you want to do this? Tuning is the way to go in my opinion.

[link Point to another website Only the registered members can access]

[Shakespear]

Predicting the dew point pressure for gas condensate reservoirs

[link Point to another website Only the registered members can access]

[link Point to another website Only the registered members can access]

[hoanglongbk88]

Thanks Shakespear for your useful advice. I feel a little bit more confident about my result I got. Can you explain more about the idea "how I show the effect of the errors on my simulation" as you said? I dont know much about simulation because my name of disser is just to make PVT model.

[hoanglongbk88]

Thank you for your advice. I feel a lit bit more confident about my result I got. Can you explain more about the idea" How I show the effect of error on simulation" as you said? I don't know much about simulation because my name of disser is just to make PVT model.

[Shakespear]

"I don't know much about simulation because my name of disser is just to make PVT model. "
oooo, if that is the situation then I would make you work harder. In other words, as your adviser I would ask that you put in more effort to get the best results and graphically prove to me that you found the best fit

I would not be easy on you, as no one does a dissertation in one month.

OK, in a very simple model, Darcy's Eq., if you knew that you viscosity could be off by +- 15 %, you would put this variation into the equation and see how much your flow would change. Hence, you could estimate how much off you would be in estimating the flow rate if your only error was in estimating the viscosity. With more parameters you could use Monte Carlo Simulation to quantify probabilities in your errors.

For a simulation this would something like this

[link Point to another website Only the registered members can access]

[link Point to another website Only the registered members can access]

[Shakespear]

I do not as a matter of fact read manuals front to back, hence was not aware that this option exists in PVTi. PVTi has a batch mode option. Something like running ECLIPSE 100 for instance.

There is a data deck that needs to be set up to run it which is explained in the manual. What does this option give you? It allows you to try different combinations of what ever you are trying to do in one shot, regression, experiments, etc.

However getting your head around the key words etc. may take a bit of time

[hoanglongbk88]

I have sent you the link 4shared to load the data I set up already, you just run file.DATA in E300. It takes about a half of hour to complete the running. After you see the results, can you answer my questions I sent yo you by email? I am waitting for your reply.

[Shakespear]

When I plot FOPR, I see the the FOPR line with the case 1(dewpoint 8000psia) is below the FOPR line of the case2( dewpoint 7150 psia). the reservior pressure is 8100 psia. I can't explain why? I think in case 1, reservior pressure will hit the dewpoint earlier than case 2. so that the percentage of liquid produces more. But in the plot it is inverse. what do you think?

First, what is flowing to the well through the reservoir rock as you being producing above the dew point (forget about the pressure drop close to the well bore)? What enters through the perf? What will you see at surface?

Next, what happens if the reservoir drops below dew point, what flows through the reservoir rock to the well ? What enters through the perf? What shows up at the surface?

Do you see something here that would account for different flow rates.

Case 1 Case 2
Pres = 8100 8100
Pdew= 8000 7150

[munqith]

Dear friends
Please i need this paper
"Gao, G., “A Simple Correlation to Evaluate Binary Interaction Parameters of the Peng-Robinson Equation of State,” Fluid Phase Equil. 74, 85, (1992)"
Thank you

[paolomaldini]

Thanks

[mirhossein]

hello dear friends,
I am working on a field's pvt. for that, I use software of PVTi (Eclipse). I followed the steps of manual for getting a match between Observes and Calculated Data. However, It is too complicated. I was wondering to know whether there are any referrences that discuss its steps clearly or not.
thank you all,