Paper Information

Title: 

COMPARISON OF ARTIFICIAL NEURAL NETWORK AND REGRESSION METHODS FOR THE PREDICTION OF THE FORMATION PERMEABILITY

Type: PAPER
Author(s): MOHAMMADPOUR IRAJ,REZAEI MOHAMMAD REZA
 
 
 
Name of Seminar: PROCEEDING OF IRANIAN MINING ENGINEERING CONFERENCE
Type of Seminar:  CONFERENCE
Sponsor:  IRANIAN SOCIETY OF MINING ENGINEERING
Date:  2005Volume 1
 
 
Abstract: 

Permeability is one of the important characteristics of the hydrocarbon formations. In fact, to determine the accurate values of permeability is an efficient and an important tool for petroleum engineers in a field’s production and management process. Many attempts have been made to obtain permeability values using laboratory measurements on core or well testing interpretation. Of course these methods are accurate but not enough to completely describe the reservoir description. Only limited wells have core or well testing interpretation, because these methods are time consuming and expensive. An objective of studying petrophysics is to provide detailed and accurate estimates of permeability in wells where no core measurements are available. This objective is difficult to achieve because no log has yet been developed that directly determines permeability in the borehole. In this study, using Regression Analysis and Artificial Neural Networks methods, permeability calculated from log data in carbonate rocks. Comparing the results of this study shows that Regression Analysis and Artificial Neural Networks methods can estimate permeability at acceptable level and Regression Analysis method performs better than Artificial Neural Networks method in predicting permeability.

 
Keyword(s): 
 
Yearly Visit 29   tarjomyar
 
Latest on Blog
Enter SID Blog