Application of machine learning algorithms in classification the flow units of the Kazhdumi reservoir in one of the oil fields in southwest of Iran

Abstract By determining the hydraulic flow units (HFUs) in reservoir rock and examining distribution of porosity permeability variables, it is possible to identify areas with suitable quality. In conventional methods, HFUs are determined using core data. This while considering non-continuity data along well, there a great uncertainty generalizing their results entire depth reservoir. Therefore, related wireline logs as continuous artificial intelligence methods can be an acceptable alternative. this study, first, number was including Winland R35, zone index, discrete type k-means. After that, by petrophysical machine learning algorithms support vector (SVM), neural network (ANN), LogitBoost (LB), random forest (RF), logistic regression (LR), have been determined. The innovation article use different intelligent comparing these each other such way that instead only two parameters permeability, obtained from logging used. increases accuracy speed reaching solution main application methodology introduced study. Mentioned compared accuracy, show SVM, ANN, RF, LB, LR 90.46%, 88.12%, 91.87%, 94.84%, 91.56% classified respectively.