Page 1 of 2 1 2 LastLast
Results 1 to 12 of 24

Thread: Use of Permeabilty in Simulation Model

  1. #1

    Use of Permeabilty in Simulation Model

    Dear Memebers,

    1. How lab generated air permeabilty is used in simulation model? What correction are required?
    2. If lab generated perm are not available. How permeability are generated?
    3. How perm is calculated from log?

    Thanks


    See More: Use of Permeabilty in Simulation Model

  2. # ADS
    Spons Circuit
    Join Date
    Always
    Posts
    Many
     
  3. #2

    Join Date
    Nov 2008
    Location
    Australia
    Posts
    508
    1. Needs to be corrected to reservoir conditions including
    (a) Klinkenberg correction (removing gas wall slip effect which tends to overestimate Darcy perm with zero wall slip assumption)
    (b) Overburden correction (squishing the rock reduces the perm to some extent - depending on the rock and depth this may or may not be significant)
    (c) Rel perm - ensure you assign appropriate rel perm. I have seen cases in tight gas wells, where the rel perm to gas at residual water saturation is 2 orders of magnitude less than air perm.

    2. SCAL, RCA, Logs, Well Tests and production history

    3. Petrophysical black magic ;-)
    Many different ways, all needing to be tuned to some extent, from simple gamma ray calculation based upon analog well data through to NMR tests inferring directly the pore size distribution, and in turn estimating resultant permeability.... it's a complex topic that I don't pretend to understand in detail. I do know though that unless you have been able to ground truth log inferred perm to RCA/SCAL/Well Test etc, that it is prone to significant errors.

  4.    Sponsored Links



    -

  5. #3
    interesting .......

    for air perm you sould correct it to klinkinberg effect
    if no liquid perm to use them to correct them you should use the klinkinberg data
    kair kliq
    0.5 0.33
    1 0.68
    10 7.8
    100 88
    1000 950


    if no core data
    there are number of correlations as textier, coates, etc the perm gonna be function of porosity
    and connate water saturation (literature)
    or you can use the pwf data to find you perm (RTA, rate transient analysis)
    if you have well test data you can use them

    from log i guess it is same correlation but what is more can give you the observed performance (history matchnig)


    i hope this is helping


    cheeeeeeeers
    Ahmed Al-Alawi

  6. #4
    You will find a good discussion of Klinkenberg effect correction here,

    [link Point to another website Only the registered members can access]
    Just reading the abstract will tell you a lot


  7. #5

  8.    Spons.


  9. #6
    Also look at this plus the directory which is public :-)

    Thank God there are some educators that SHARE their material with the World !!!! (I took his course)

    There is a LOT of EXCEL spreadsheets here

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

    Last edited by Shakespear; 09-23-2010 at 07:56 PM.

  10. #7
    We have a field which is recently discovered after drilling one exploratory well
    1. The well is producing from particular sand during last 8 months @ 40 m3/d on self.
    2. This is only one well is producing from this field
    3. Core analysis of 7 samples are available for this sand
    4. Build up study was done, permeability is 9 mD.
    5. No log derived perm is available

    Core Analysis Data
    Phi Ka,md(klinkenberg corrected)
    0.25 59.4
    0.22 109.3
    0.22 133.8
    0.20 128.9
    0.20 75.3
    0.22 85.1
    0.20 38.3


    From rel-perm study, Ko at Swi= 21 mD
    Now my question is that if I go for simulation study for this field taking other parameters, how I do permeability modeling of this sand ?

    Thanks
    Last edited by dipak_m; 09-23-2010 at 08:53 PM.

  11. #8
    kair kliq
    0.5 0.33
    1 0.68
    10 7.8
    100 88
    1000 950


    opn excel sheet x axis kair and yaxis kliq
    then make trend line linear use the equation to convert your air perm to liq or klinkinberg

    as simple as this
    i hope it is clear ...............

  12. #9

  13. #10

    Join Date
    Nov 2008
    Location
    Australia
    Posts
    508
    The buildup (assuming it was interpreted correctly) ground truth's your bulk reservoir average perm for an assumed/estimated h. Was it over the entire zone that was cored?

    The core analysis, while interesting, has to be looked at in the context of where those plugs were taken - match up the depths of the core samples to your logs (gamma / resistivity / ND crossover etc) and try to get a feel for how much of each sort of rock is in your pay zone. Do you have representative population of samples tested (often people will fall into trap of only testing best looking bits, meaning the majority of your more rubbish reservoir is uncharacterized)

    With your rel perm Kro at Swi of 21mD - which sample was this against? What was the Kair? How much reduction from Kair/Kro was there? Is it the same rock type as the rest of your samples?

    When all is said and done, YOU are the one that has to make a judgement call on how to distribute/represent the perm. Going by the DST result, you will need a significant fraction of poorer rock through the bulk reservoir.

    Also remember that about the only thing that is 100% certain is that the rock you have sampled is no longer in the reservoir......

  14. #11
    Also remember that about the only thing that is 100% certain is that the rock you have sampled is no longer in the reservoir......
    Good One, never looked at it this way :-)

    Please look at this basic explanation of the Klinkenberg effect correction

    [link Point to another website Only the registered members can access]
    Crane is good !!!

    A bit more information with what you are dealing

    [link Point to another website Only the registered members can access]
    Other good info is found on the SPE Queensland site here

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

    Last edited by Shakespear; 09-24-2010 at 10:32 AM.

  15.    Spons.


  16. #12
    I just looked at Crane's spreadsheets and perhaps this one may help you

    4. META/CORE Core Analysis, includes net pay, productivity, and reserves for comparison with META/ESP and META/TAR, plus crossplots.
    English and Metric Units

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


  •   

Similar Threads

  1. Simulation model QC
    By pet_engi in forum Reservoir
    Replies: 3
    Last Post: 02-13-2014, 11:13 AM
  2. Replies: 2
    Last Post: 05-08-2013, 08:48 AM
  3. Simulation model initialization & Oil in place
    By dquento in forum Reservoir
    Replies: 2
    Last Post: 02-25-2013, 02:15 PM
  4. Pressure in the simulation model
    By bayan in forum Reservoir
    Replies: 3
    Last Post: 12-11-2012, 04:31 PM
  5. Please help in simulation model
    By bayan in forum Reservoir
    Replies: 4
    Last Post: 12-11-2012, 11:02 AM

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •