Paper Information

Title: 

PERMEABILITY DETERMINATION IN PARSI OIL FIELD RESERVOIR USING ARTIFICIAL NEURAL NETWORK

Type: PAPER
Author(s): TADAYONI M.,RABANI A.R.,NABI BID HENDI M.
 
 
 
Name of Seminar: PROCEEDING OF IRANIAN MINING ENGINEERING CONFERENCE
Type of Seminar:  CONFERENCE
Sponsor:  IRANIAN SOCIETY OF MINING ENGINEERING
Date:  2005Volume 1
 
 
Abstract: 

Formation permeability is either measured in the laboratory from cores or evaluated from well test data. Asmari formation is the most important pay zone in Parsi oil field. In this formation, lithology and depositional environments are very complex, and there is not enough core and well data. These difficulties make it unable to determine permeability with fair precision. In this study, an artificial neural network (ANN) has been designed to predict the permeability of the formation using the well log data in Parsi field. The study field is located in south-west of Iran. The ANN for permeability uses gamma ray, density, sonic, neutron logs in addition to depth in three wells for training, testing and validation processes.
In generalization process, the obtained correlation coefficient between ANN permeability and core permeability is 0.689. Because of poor precision of network in generalization process, initially Parsi field is devided into five zones and then training, testing and validation processes are performed in the first four zones. Ultimately in Generalization process the correlation coefficient between ANN permeability and core permeability in zones 1, 2, 3 and 4 are obtained equal to 0.932, 0.961, 0.887 and 0.921, respectively.

 
Keyword(s): 
 
Yearly Visit 5   tarjomyar
 
Latest on Blog
Enter SID Blog