AI based mechanistic modeling and probabilistic forecasting of hybrid low salinity chemical flooding

摘要: Abstract Over the past decades, it has been widely shown that Low Salinity Waterflooding (LSW) outperformed High (HSW) in terms of higher oil recovery, particularly combining with other conventional Enhanced Oil Recovery (EOR) methods such as chemical flooding to benefit from their synergies. This paper presents a novel approach mechanistically model Hybrid Chemical Flooding, with: (1) development hybrid EOR concept decades; (2) utilizing Multilayer Neural Network (ML-NN) artificial intelligent technique robust Equation-of-State reservoir simulator fully coupled geochemistry; (3) systematic validation laboratory data; and (4) uncertainty assessment LSW process at field scale. Various parameters polymer, surfactant, salinity can affect on relative permeability simultaneously during recovery processes. To overcome this problem, ML-NN was applied for multidimensional interpolation permeability. Additionally, used within Bayesian workflow capture uncertainties both history matching forecasting stages The proposed indicated good agreements various coreflooding experiments including HSW, LSW, Surfactant (LSS), where efficiently complex geochemistry, wettability alteration, microemulsion phase behavior, synergies occurring these