Loading...

Sensitivity Analysis in Total Search Domain and Automatic Selection of the Most Sensitive Matching Parameters in Tuning of EOS and History Matching

Chenani, Ahmad | 2013

769 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 44711 (06)
  4. University: Sharif University of Technology
  5. Department: Chemical and Petroleum Engineering
  6. Advisor(s): Jamshidi, Saeed
  7. Abstract:
  8. One of the major issues raised in the numerical models is the effects of various input parameters on the model output. Identifying the most important parameters in the output is very important in petroleum engineering, especially in tuning equation of states and history matching. Screening the most important parameters from unimportant parameters is assessable by sensitivity analysis.In this study local and global sensitivity analysis methods are used to identify the most important parameters in tuning EOS for seven different fluid samples. Global sensitivity analysis methods are more successful in correct ranking of tuning parameters. Morris method as a global method can screen the important parameters from unimportant one’s. However it can’t rank the important parameters precisely.Sobol method is a precise method for ranking tuning parameters based on their effects on results of the simulations using EOS’s. However this method is numerically expensive and it’s not efficient to use it with high number of tuning parameters. A two stage sensitivity analysis method is proposed in this study which is a combination of Morris and Sobol method. Using sensitivity analysis method is very important in identifying most important parameters in reservoir simulation results, prior to history matching. Due to high number of function calls and time consuming calls of reservoir models, it’s not feasible to use Sobol method for Sensitivity Analysis in real reservoir models. Comparison between the results of local differential analysis and global Morris method shows that Morris method is more successful in distinguishing important parameters from unimportant parameters in reservoir simulation results
  9. Keywords:
  10. History Matching ; Sensitivity Analysis ; Tuning Equation of States ; Morris Method ; Sobol Method

 Digital Object List

 Bookmark