m tadayoni - corporate planning directorate NIOC
h yadegari - international affairs and technological devision RIPI
nasser keshavarz faraj khah - center for exploration and production RIPI

porosity and water saturation of oil reservoir rocks are usually determined by core analysis and well test methods . However these methods are expensive and time consuming . Also because of lithology changes heterogeneity of reservoir rock and nonexistence of sufficint well cores determination of the parameters by these usual methods are not accurate . so the best way to decrease cost , increase accuracy , and decrease time in applying advanced soft ware such as geolog and back - propagation artificial neural network (BP-ANN). IN THIS PAPER , A BP-ANN is designed to predict the porosity and water saturation of formations using the well log data in X oil field located in south-west of iran. the data of one well No.19 that has core data is used for training , testing validation and generalization processes. then the BP-ANN results are compared to the results obtained by geolog software. with respect to the results it is concluded that the BP-ANN is more accurate than geolog software in determining porosity and water saturation. finally water saturation and porosity are simulated in three other wells Nos48,49,and 64 that do not have core data.

geolog soft ware , back propagation artificial neural network , porosity , water saturation.