Sensitivity Analysis of Fluid–Fluid Interfacial Area, Phase Saturation and Phase Connectivity on Relative Permeability Estimation Using Machine Learning Algorithms

Recent studies have shown that relative permeability can be modeled as a state function which is independent of flow direction and dependent upon phase saturation (S), connectivity (X), fluid–fluid interfacial area (A). This study evaluates the impact each three parameters (S, X, A) in estimation permeability. The importance four separate quadrants S-X-A space was evaluated using machine learning algorithm (out-of-bag predictor method). results show sensitive to all parameters, S, A, with varying magnitudes at constant value wettability. We observe wetting-phase most saturation, while non-wetting connectivity. Although least important, still important make more exact function.