Mathematical Prediction of Oil Recovery Factor for Nanoparticles Assisted Polymer Flooding through Statistical Analysis and ANN Modelling

Conventional Polymer flooding is a promised CEOR technique.However, there are some shortages associated with molecular architecture destruction in severe reservoir environments.The grafting of nanoparticles (NPs) toward polymer macromolecule may ameliorate the flooding performance and hence increase the oil recovery further than conventional polymer flooding.The potential of NPs assisted polymer flooding has been extensively studied in porous media through core flooding.The target of this work is to statistically scrutinize the previous core flooding experiments and investigate the parameters that affect the incremental oil recovery by NPs assisted polymer flooding.More than 350 core-flood tests were gathered to investigate the impacts of rock-oil properties, flooding conditions, polymer, and NPs characteristics on incremental oil recovery.Furthermore, an empirical model was built using an Artificial neural network (ANN) to predict the incremental oil recovery for polymeric nanofluid.The results of the investigation indicated that NPs assisted polymer flooding with optimum concentration of both polymer and NPs may result in an incremental recovery of 18% of oil-in-place.In addition, the results revealed that core-flooded viscous oil needs high permeable rock to guarantee successful flooding.Coefficient of determination (R 2 ) between the real and calculated incremental oil from the ANN model was established to be 0.953 and 0.952 with an error of 5.6 and 8.7% respectively for the training and testing approaches.Such statistical investigations and ANN approaches afford new insights and guidelines for preliminary assessment, designing, and execution Nanohybrid polymer upcoming projects in field scale.