Evaluation and prediction of Enhanced Oil Recovery (EOR) by Low Salinity Water Flooding (LSWF) injection

Abstract

Low Salinity WaterFlood (LSWF) injection is an Enhanced Oil Recovery (EOR) method proven to be effective by extensive experimental studies. Correct implementation of this method in reservoir-scale simulations requires reliable estimation of relative permeability data. However, due to the ongoing debate regarding the dominant mechanism in this process and the inadequate understanding of complicated brine-oil-rock interactions, only a few models have been suggested to estimate relative permeability associated with LSWF for either carbonate or sandstone rocks. Existing models simulate the impact of LSWF based on geochemical interactions; however, the fluid-fluid interaction has been overlooked. Some models depend on the cation-exchange capacity of clay which is not adequate for clay-free rock. In contrast, others are based on weighting factors such as the divalent ions desorption, which is case dependent. This study presents a novel semi-empirical model of LSWF relative permeability, after high salinity water flooding, based on incremental oil recovery measured during low salinity injection. Therefore, it can be applied to all rock types, fluid systems, and wettability conditions regardless of the active mechanism. The relative permeability at the high salinity water flooding, krHS and incremental oil recovery during low salinity injection were inputs to the model for predicting the low salinity kr curve, krLS and, consequently, this model can be used to assess the performance of LSWF. Well-known mechanisms in literature on the incremental oil recovery during LSWF are reviewed including micro-dispersion. In this work, correlation between the incremental oil recovery and the amount of micro-dispersion has been employed for sensitivity analysis, and evaluation of the new model's response among various levels of incremental oil recovery. This new model has been validated utilising 12 coreflood datasets obtained from core flooding experiments under both unsteady-state and steady-state flooding conditions or protocols. A new steady-state experiment has been performed in this study to produce the first reliable relative permeability data that is needed for validation. This dataset, along with the unsteady-state dataset obtained under tertiary and secondary mode by other researchers in our group at Heriot-Watt University, has been used to validate the new model and to quantify the effect of LSWF injection on the relative permeability. Five experiments have already been published in literature by other researchers from our group, while the other four experiments have not yet been published. Additional two more experimental data from other group researchers available in the literature were deployed for further verification. Due to its mechanism independence, the new model suggested in this thesis can be applied for efficient performance screening of all LSWF injection scenarios, which is invaluable for the oil and gas industry’s decision-making process. The results confirm the difference in relative permeability curves between high-salinity and low-salinity injections caused a decrease in water relative permeability and an increase in the oil relative permeability. They also prove that low-salinity brine can shift the rock wettability from oil-wet or mixed-wet towards a more water-wet condition. The obtained relative permeability curves extend across a substantial saturation range, making this valuable information necessary for numerical simulations. To the best of our knowledge, the data from the steady-state experiment is a first in assessing the impact of low-salinity waterflooding at a steady-state condition using a reservoir live crude oil and long reservoir core sample at reservoir condition. The results of this study are of utmost importance for the oil and gas industry.