A New Approach to Simulate Low Salinity Water Flooding in Carbonate Reservoir

Abstract

Abstract Low salinity water flooding (LSWF) as an enhanced oil recovery method (EOR) has attracted increased attention from oil companies due to its numerous benefits and advantages. It has been confirmed in several studies and laboratory experiments that LSWF has improved oil recovery. However, the underlying mechanism responsible for such an impact is still debatable. All previous studies focused on a geochemical process where fluid-fluid interaction has been overlooked. Recently, some studies have indicated that brine-crude oil (micro-dispersion) interactions play dominant roles in improved oil recovery in carbonate rocks. Nevertheless, at the moment, no commercial simulator can mimic this mechanism from the perspective of fluid interactions. In this work, we investigated whether micro-dispersion is applicable in commercial reservoir simulators through the history matching of two carbonate coreflood experiments. In this part of the investigation, three aspects will be addressed. (i) Develop a correlation of the link between the mechanism (micro-dispersion) in the lab and numerical simulation. (ii) Predict the low salinity relative permeability curves. (iii) History match the experimental data. This paper presents an integrated method of simulating low salinity water floods in carbonate rocks. Two different approaches have been applied to the history matching of unsteady state coreflood experiments. First, numerical simulation was performed to extract the high salinity relative permeability curves (KrHS) of the secondary mode for both experiments. Then, the findings from the first approach and the experimental results were used to develop a new approach for predicting the low salinity relative permeability (KrLS) curves. The new approach was not only used to predict KrLs curves through micro-dispersion but also used as input to history match the tertiary low salinity water floods. An excellent match was obtained using both the numerical simulation model and the new approach for the oil recovery and pressure drop profile, where two different relative permeability sets were generated in this study for each coreflood. The first observation promotes the premise that a history match of a coreflood can be obtained using different sets of relative permeability curves. In contrast, the Corey exponents, residual oil saturations and endpoints are essential parameters in the history matching of LSWF. The results obtained in this study will help to understand the modelling process involved during oil recovery by LSWF and introduce a new approach to model the effect of LSWF.