سال انتشار: ۱۳۸۵
محل انتشار: اولین کنگره مهندسی نفت ایران
تعداد صفحات: ۱۲
Ali Kadkhodaie Ilkhchi – P.H.D student in petroleum Geology, School of Geology, University of Tehran, Tehran Iran
Mohammadreza Rezaee – Associate Professor of Petroleum Geology, School of Geology, University of Tehran, Tehran, Iran
Seyed Ali Moallemi – Head of Petroleum Geology Department, Research Institute of Petroleum Industry, Teran, iran
Nilofar Masoodi – MSc.Student in peroleum Geology Olom va Tahghighat Branch, Islamic Azad University, Tehran, Iran
pereability and rock types are the most important rock properties which can be used as input parameters to build 3D petrophysical models of hydrocarbon reservoir. As well as, log data of prime importance in acquring petrophysical data from hydrocabon reservoir. Reliable log analysis in the hydrocarbon resevoir requires a compelet set of logs. For any number of reasons such as incomplete logging in old wells, destruction of logas due to inappropriate data storage, and measurement errors due to logging toolproblems hole conditions, log suites are either incomplete or not reliable. In this study, a fuzzy c-means (FCM) clustering technique was use to rock types classification based on porosity and permeability data. Then base on fuzzy inference system. Then a back propagationneural network with trainlm training function was applied to verify fuzzy results for permeability modeling. For this purpose, two wells of the Southern Iran Fields were chosen to construct intelligent models of the reservoir and a third well was used as a test well to evaluate the reliability of the models. The results of this study showed that fuzzy logic approach was successful for prediction of well logs, permeability and rock type in the studied reservoirs.