Saturation height function

[dipak_m]

Well Method Core_ID Interval Plug_no Phi Ka Sw Pc
(%) (mD) (%) (kg/cm2)
XYZ centrifuge CC-1 2854.5-2862.5 8709 15.28 47 0.99 0.06
0.52 0.49
0.26 0.97
0.23 1.79
0.20 3.13
0.18 6.44
0.12 14.06

XYZ centrifuge CC-1 2854.5-2862.5 8705 13.93 35 0.99 0.05
0.59 0.45
0.30 0.88
0.25 1.63
0.21 2.86
0.18 5.87
0.12 14.06

XYZ centrifuge CC-1 2854.5-2862.5 8708 14.57 38 0.97 0.05
0.51 0.47
0.24 0.91
0.19 15.05
0.14 2.96
0.13 6.07
0.09 14.06

XYZ centrifuge CC-1 2854.5-2862.5 8703 11.76 3.3 0.96 0.05
0.87 0.49
0.66 0.95
0.48 1.76
0.33 3.08
0.23 6.32
0.17 14.06

[dipak_m]

Capillary presure Data

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

[swibow]

Guys, sorry for interupting this excellent discussion,

I got a saturation equation from a petrophysicist, but I don't really understand. Here is the equation:
swht = limit(.44/(.00766*(ht*.21)*((kpor/port)**.5))**.212, .15, 1)

Is there anyone who can give me a clue regarding the above equation?
Especially what is "limit" and the "double star" (**) mean?

Thanks a ton for your help!

[temr]

Dear Temr,
There is an initialization chapter in your thread. But I want in details. What are the different methods of initialization. I want to know every details of each initialization process, from Xls file to Petrel and then Eclipse.
Thanks

Dipak ,
I will repeat what has been already written
1. Calculation of block depth ( center point of the block)
2. Calculation of block rock pressure
3. Calculation of Capillary pressure of phases
4. Calculation of block water saturation based on capillary pressure
5. The function of Sw and Pc are comes from core/log data
Everything this is detail described in th chapter
And this is basics which a must to understand the saturation modeling in simulator
If you want more then use manual and read extra books
Regards
A

[temr]

Guys, sorry for interupting this excellent discussion,

I got a saturation equation from a petrophysicist, but I don't really understand. Here is the equation:
swht = limit(.44/(.00766*(ht*.21)*((kpor/port)**.5))**.212, .15, 1)

Is there anyone who can give me a clue regarding the above equation?
Especially what is "limit" and the "double star" (**) mean?

Thanks a ton for your help!

May be you need to talk to petrophysicis)?
It is kind of J function with poro and perm and may be Owc column high but please check what this parameters are intend to and use them in petrel or eclipse to calculate you sw

[vinomarky]

OK - have had a quick/rough look at the data, and uploaded a sheet at the following address *updated*

[link Point to another website Only the registered members can access]
Assumptions:
- Oil/Brine centrifuge test
- Oil/Brine reservoir
- Reservoir oil at 0.825 SG in-situ
- Reservoir brine 1.02 SG in-situ

Notes:
You will never get perfect fits across entire population. Need to critically decide which areas are more important to get better fit it. If your column is only 100ft high, then if the 500ft+ data does not fit so well, remove it from the fit criteria. If you have 99% of your reservoir with Sw<0.7, then don't worry so much if your relationship has a poorer fit with high water saturations. Of course we should strive to get the best fit we can, but in the end you will in all likelihood need to make compromises to better address those areas that are most important

With J-Function, I assumed max hydrocarbon column of 600ft, so did not give any weight to errors in the 1,500ft points

All samples seem to be exhibiting 'normal' type trends - insofar as I'd probably lump them as one saturation function family unless there was some other data to tell me otherwise

Be careful working with Ka (perm to air). Often in situ perm can be an order of magnitude less through overburden correction and rel perm to the primary reservoir fluid flowing. Ensure that you are consistent with how you test/populate/relate all the various perms (ie lab test perm, property model perm, rel perm) so that the final property perm x rel perm reflects in situ reservoir deliverability for your fluid and that your saturation function is used against the appropriate perm value. I'm not saying you need to always work with Ka, or always with Perm to hydrocarbon or water - what I am saying is be consistent through your workflow.


Fit Results (updated):
J = 0.4778 * Sw^-2.356, using Perm exponent of 0.529 and Poro exponent of 0.608 (J = Pc x (Perm^0.529)/(Poro^0.608)/26 )

or if solving for Sw given a calculated J;
Sw = 10^( (-log(J) + log(0.4778))/2.356)

Lambda constants are;
a1 39.08174161
a2 -14.76172542
l1 0.726524565
l2 0.072382779
b1 0.17321564
b2 -0.148686704

For 'no name' relationship;
Choosing 675ft as Swir -
Swir = 0.2581 - 0.06106 x Log(K)
Alpha 1324253.612
Beta 33922.95145

Comparing the results, it appears that both the J-function as well as Lambda functions give very good matches based upon the data furnished to date. This being the case, I'd probably elect to use J-function due to ease of implementation in Eclipse.

With regards to the petrophysics equation - yes, go ask them. If I had to guess, I would translate it as follows;
Swi = .44/(.00766*(Height*.21)*(SQRT(PERM/PHIT))^.212, but not allowed to go below 0.15 or above 1.0

[swibow]

Aha !! I got the xls file, Thanks vinomarky!!

About the equation, I have the same guess with you. Unfortunately when I try to recalculate in my model, the result is totally different.
There is no petrophysicist here, and the one who made the equation has gone. Anyway thank you very much temr and vinomarky.
I love this forum.

[vinomarky]

Check (a) what FWL they are using and (b) whether they are using m or ft - two obvious areas for significant differences

Also, there are a few ways petrophysicists calculate Sw - ensure you are benchmarking against the same method

[swibow]

Check (a) what FWL they are using and (b) whether they are using m or ft - two obvious areas for significant differences

Also, there are a few ways petrophysicists calculate Sw - ensure you are benchmarking against the same method

Yes you are right. Problem's solved. He used meter instead of ft, don't know why.

Vinomarky, I have downloaded the solver add ins, but I don't know how to use it, could you give an example how to calculate one variabel like a1 ? Pleaseeee......

[vinomarky]

Go to Lambda tab, change constants in cells T7-T12 to 1.0 (something different from the current fit)

Tools - Solver
Set target cell - T14
Equal to: min
By changing cells: T7:T12
Solve

[swibow]

Go to Lambda tab, change constants in cells T7-T12 to 1.0 (something different from the current fit)

Tools - Solver
Set target cell - T14
Equal to: min
By changing cells: T7:T12
Solve

That's it?? That easy?
Thanks thanks thanks .....Vinomarky. I learn a lot from you today.

[reservoir_engineer]

many thanks Vinomrky,
you are great friend and we learn too much from you.
regarding flow units definitions, i got paper from friend two years ago.
i hope it help,