边界层对致密油藏逆向渗吸的数值模拟

致密油藏开发过程中一般采用水平井多段体积压裂方法进行开采,近井基质被缝网充分切割,极大的增加了基质与裂缝的接触面积,压裂液或水通过缝网与基质发生逆向渗吸,因此无论压裂过程还是注水吞吐开发过程都不能忽视逆向渗吸作用的影响,另外由于致密储层流动规律不同于普通油藏,这导致致密油藏的逆向渗吸与常规油藏的逆向渗吸作用存在许多不同,为了准确模拟裂缝性致密储层中的渗吸作用,首先采用耗散粒子动力学的方法验证了致密储层微纳米喉道中渗吸过程中边界层的存在并得到渗吸过程中边界层厚度的变化规律,然后利用孔隙网络模型计算考虑边界层的关键渗流参数,主要包括渗透率、毛管力曲线及相渗曲线;并将计算获得的关键渗流参数应用到表征致密储层复杂缝网的数值模拟模型中,模拟结果显示边界层对渗吸作用具有抑制作用,当考虑边界层后储层的渗吸能力大幅度减弱,致密储层的渗吸采收率可能仅为不考率渗吸时的3/8而已,因此应当重视致密储层中边界层的存在,否则可能会高估储层的流动能力。

Numerical simulation on tight oil reservoir reverse imbibition with boundary layers

Multi-stage volume fracturing of horizontal wells is common in the development of tight oil reservoirs,where the matrix near wellbore is fully cut by and in large-scale contact with the fracture network,resulting in reverse imbibition of fracturing fluid or water into the matrix through the fracture network.Therefore,the reverse imbibition effect shouldn't be ignored in both fracturing and water-flooding.In addition,due to the difference of flow pattern between tight and ordinary reservoirs,the factors contributing to the reverse imbibition effects are accordingly different in the two distinct reservoirs.To accurately simulate the imbibition in fractured tight oil reservoirs,we apply the Dissipative Particle Dynamic(DPD)method to verify the existence of boundary layers during the imbibition in nano-scale pore throats in tight reservoirs,and discover the variation pattern of the boundary layer thickness;then calculate the critical seepage parameters with boundary layers considered via the pore network model,mainly including permeability,capillary pressure curve and relative permeability curve;and finally incorporate these parameters into the numerical simulation model to characterize complex fracture networks in tight reservoirs.The simulation results show that the boundary layers serve to greatly reduce imbibitions-the imbibition recovery in tight reservoirs with boundary layers is only 3/8 of that without considering imbibition.In conclusion,the boundary layer in tight reservoirs deserves great attention;otherwise,the flow potential of these reservoirs may be overestimated.