علی رنجبر

The shale volume in the formation affects the petrophysical parameters of the reservoir, such as porosity, permeability, and saturation. Since these parameters indicate the amount of hydrocarbon in the reservoir and the reservoir producibility, shale volume determination is an essential challenge in the oil industry. This study compares conventional methods and Machine Learning (ML) algorithms to estimate the shale volume in the Fahliyan Formation in one of the oil fields in southern Iran. In this regard, the shale volume of the formation was calculated using Gamma-Ray (GR), Density-Neutron (DN), and Density-Sonic (DS) as well as ML methods such as Artificial Neural Networks (ANN), Support Vector Machine (SVM), and Random Forest (RF). SP, GR, CGR, HLLS, HLLD, DT, NPHI, RHOZ, and PEFZ logs are used as input data in these methods. The results show the low performance of DN and DS methods in estimating shale volume. The coefficient of correlation (R2) for these methods are 0.66 and 0.4, respectively, and the Root Mean Square Error (RMSE) is calculated 0.01 and 0.02, respectively. However, ANN, SVM, and RF methods estimated the shale volume with much better performance. R2 for these methods are 0.97, 0.89, and 0.91, respectively, and RMSE is calculated 0.0008, 0.0031, and 0.0035 for them, respectively.