Abstract Emulsion flow in porous media is of paramount importance to the use of emulsions in the conformance control and enhanced oil recovery processes. In this paper, a new theoretical model, incorporating physical properties of porous media, physicochemical properties of the emulsion system, injection strategy, and the interactions between porous media and emulsion, was developed to quantitatively describe flow behaviors of emulsions in porous media. The resistance factor of an emulsion when transported in porous media was first derived through a-two phase flow method. The strong interaction between emulsion droplets and porous media was characterized by the capillary resistance force in the model. A non-uniform capillary model which considers size differences of the pore-body and pore-throat in porous media was proposed to represent the complicated real porous media. By analyzing the adsorption and plugging properties of different emulsion droplets in the non-uniform capillary model, the capillary resistance force was finally determined. To describe emulsion flow in the subsequent water flooding process after emulsion injection, an emulsion dilution factor was introduced into the model. A set of experimental data of emulsion flow in sandpacks was used to validate the reliability of the newly proposed model. The validation results show that by appropriately choosing coefficients, the simulated results are in good agreement with experimental values, with a maximum average absolute error less than 10%, and the developed theoretical flow model can be used to describe emulsion flow behavior in porous media.

中文翻译：

---

用于一致性控制的多孔介质乳液流动新模型

摘要 多孔介质中的乳液流动对于乳液在一致性控制和提高采收率过程中的使用至关重要。在本文中，开发了一种新的理论模型，结合多孔介质的物理性质、乳液体系的物理化学性质、注入策略以及多孔介质与乳液之间的相互作用，以定量描述乳液在多孔介质中的流动行为。乳液在多孔介质中传输时的阻力系数首先是通过两相流法推导出来的。乳液液滴与多孔介质之间的强相互作用通过模型中的毛细管阻力来表征。提出了一种考虑多孔介质中孔体和孔喉尺寸差异的非均匀毛细管模型来表示复杂的真实多孔介质。通过分析不同乳液液滴在非均匀毛细管模型中的吸附和堵塞特性，最终确定了毛细管阻力。为了描述乳液注入后后续水驱过程中的乳液流动，在模型中引入了乳液稀释因子。一组砂包中乳液流动的实验数据被用来验证新提出的模型的可靠性。The validation results show that by appropriately choosing coefficients, the simulated results are in good agreement with experimental values, with a maximum average absolute error less than 10%,