深地含不同流体岩石破裂机制的离散元流固耦合模拟研究

Discrete Element Fluid-Solid Coupling Simulation of Fracture Mechanisms in Deep Fluid-Bearing Rocks

摘要: 深地油气储层赋存于高地应力、高地温及高孔隙压力的复杂地质环境中，揭示含流体岩石破裂机理对储层裂缝预测与甜点识别具有重要意义。本研究基于离散元流固耦合理论，采用PFC2D数值模拟方法，针对含水、油、甲烷三种流体赋存岩石开展三轴压缩对比研究。通过构建数字岩心模型，标定岩石颗粒细观力学参数及流体(水、油、CH₄)物性参数，设置不同围压与孔隙压力实验条件，系统分析了含不同流体岩石的宏细观力学响应特征。研究结果表明：(1) 流体类型显著影响岩石抗压强度，水相介质使应力峰值降幅最大，油相介质次之，气相介质(CH₄)影响最小；(2) 围压水平调控颗粒运移机制，水的润滑作用、油的黏滞阻力及CH₄的可压缩性分别影响颗粒位移程度与速度；(3) 细观破裂模式呈现流体依赖性，含油岩石剪裂纹较多，含水与含CH₄岩石更接近脆性破裂，力链分布也因流体特性呈现不同密集程度与拉力链数量差异。研究成果揭示了多种流体–岩石相互作用机制，为深层油气储层压裂优化和甜点预测提供了理论依据。

Abstract: Deep hydrocarbon reservoirs are situated in complex geological environments characterized by high in-situ stress, elevated temperatures, and high pore pressures. Understanding the fracture mechanisms of fluid-bearing rocks is crucial for reservoir fracture prediction and sweet spot identification. This study utilizes a discrete element fluid-solid coupling approach with PFC2D numerical simulations to conduct a comparative investigation of rocks saturated with water, oil, and methane (CH₄) under triaxial compression. A digital core model was established, incorporating calibrated mesoscale mechanical parameters for rock particles and fluid-specific properties (water, oil, CH₄). Experiments were performed under varying confining and pore pressures to systematically analyze the macro- and meso-scale mechanical responses of fluid-bearing rocks. Key findings reveal that: (1) Fluid type significantly affects rock compressive strength, with water causing the greatest reduction in peak stress, followed by oil, while CH₄ exhibits the least influence; (2) Confining pressure governs particle displacement mechanisms, where the lubricating effect of water, viscous resistance of oil, and compressibility of CH₄ distinctly modulate displacement magnitude and velocity; (3) Mesoscale fracture patterns are fluid-dependent, with oil-saturated rocks developing more shear cracks, whereas water- and CH₄-bearing rocks predominantly undergo brittle tensile failure. Force chain distributions further demonstrate fluid-induced variations in network density and tensile chain proportions. These results elucidate the interaction mechanisms between multiphase fluids and rock matrices, offering theoretical foundations for optimizing hydraulic fracturing and enhancing sweet spot prediction in deep hydrocarbon reservoirs.