The major problem that concerns log analysts is their interpretation for shaly formations. The previous models that were used to estimate the hydrocarbon saturation in shaly formation revealed a wide area of discrepancies. Shaly sand models based on shale volume fraction (Vsh) fail to consistently predict representative values of hydrocarbon saturation from wire line data. Cation exchange capacity-based models require laboratory-determined concentration of sodium exchange cation associated with clay (Qv) value which is not commonly available to the log analyst. In addition, different laboratory techniques are found to yield different Qv values for the same core sample. Moreover, Louisiana State University (LSU ) model is very complex in their calculation and depends on the determination of Qv by using a complicated technique. Also, the theoretical model such as Berg's model, use a repeated complicated iteration technique. Therefore, a new shaly model is presented named ANN's model. It is built a non-linear model to predict water saturation as a function of conventional well logs. The neural network was trained and tested over a wide range of Bahariya formation, which characterized by high clay content. The resulting models provide an excellent correlation between the inputs and outputs. The accuracy of results derived from the ANN's model is highly exceeds that of the previously published shaly sand models. In general, an ANN's model displayed a correlation coefficient of 0.9968 where 1.0 is a perfect match. Moreover, maximum absolute error of less than two percent were observed.

You can access this article if you purchase or spend a download.