Page 3 of 3 FirstFirst 1 2 3
Results 25 to 29 of 29
  1. #25

    Re: History Matching with ECLIPSE

    The idea of having a single one "best" history matched method is no longer a standard practice in oil industry,Now s days people tend to have a "Group" of history matched model using the Assisted History Matched Approach.For this there are a number of different softwares availabe personaly i recommned using SPT Group "MEPO" which is very powerfull and speed thing up in history matching.
    I also read a series of SPE papaers explaning the history matching step by step published by SPE :

    [link Point to another website Only the registered members can access]
    I'm sure you can find the link on ,hope this helps you.

    See More: History Matching with ECLIPSE

  2. # ADS
    Spons Circuit
    Join Date
  3. #26

    Re: History Matching with ECLIPSE

    i wrote code to process history matching problem./ if anybody need help this problem, can contact with me. My email:

    My threads; vanthai88 :

    •    Sponsored Links


    • Re: History Matching with ECLIPSE

      If anyone needs help contact me as well

      My threads; GateKEEPER32 :

      • #28

        Re: History Matching with ECLIPSE

        I advise you code your history matching problem using EnKF instead*because you will be stuck in a local minimum. In my research I coded my history my history matching code using EnKF as specified in Evensen 2003* and compared the RMS with that from SIMopt* I foud out my EnKF inversion gave a lower RMS when compared to SIMopt* the history matched realization permeability field from EnKF gave a higher SSIM index to the true permeability maps than when compared to SIMopt.This is because SIMopt is a determinsitc method and EnKF is a stochastic method. In theory* we expect any stochastic method to be superior to the deterministic case* we pay for this in the computation time required for each realization in the ensemble.This is overcome by parallel computing(OMP) this dynamic* EnKF updates itself by the KALMAN gain matrix of the ensemble(group) which is a matrix inferred from the covariance matrix between the data(FOPR* WCT*WBHP...) and the parameter states(permeability/porosity...)

        Follow the steps in EVENSEN 2003* the FORTRAN code is in there.MRST have an equally excellent code as well. What you need to do is produce your initial ensemble using SGSIM from GSLIB or SGeMS* or for multiple point FILTERSIM from SGeMS.
        Next* choose which parameters are uncertain. It may be permeability/an or porosity...FTM etc
        Run the forward model for each realization in the ensemble
        Call the EnKF subroutine to solve the inverse problem
        Move on to the next time step.
        Last edited by Clementetienam; 12-17-2016 at 08:14 PM.

        My threads; Clementetienam :

        • #29

        •    Spons.


        Similar Threads

        1. History matching in Petrel ?
          By bratek in forum Engineering Software Tutorial
          Replies: 8
          Last Post: 01-26-2015, 03:54 PM
        2. Errors in History Matching
          By UsmanHWU in forum Reservoir
          Replies: 0
          Last Post: 05-05-2014, 10:33 AM
        3. History matching using SimOpt in Eclipse
          By wampomah in forum Reservoir
          Replies: 1
          Last Post: 08-06-2013, 01:15 AM
        4. History matching techniques
          By Veggy in forum Reservoir
          Replies: 0
          Last Post: 05-18-2012, 09:50 AM

        Tags for this Thread


        Posting Permissions

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