页岩压裂裂缝网络预测方法及应用

页岩储层天然裂缝发育，加上水力压裂所形成的复杂裂缝网络，可以为油气流动提供有效通道，然而复杂裂缝网络扩展对压裂裂缝识别与优化设计也提出了新的技术挑战。为此，综合考虑天然裂缝与水力裂缝相互作用、缝间“应力阴影”等影响，应用流体力学与断裂力学理论，建立了裂缝网络扩展模型（FNPM），实现了不同力学参数、不同天然裂缝等条件下裂缝扩展方位、几何尺寸、支撑裂缝面积等多指标预测。理论模拟和现场实例应用结果表明：①改造裂缝网络扩展受控因素多，地应力差、簇间距及净压力等参数决定了“应力阴影”效应的强弱，天然裂缝与水力裂缝之间的相互作用机理决定了改造裂缝网络的复杂程度；②地应力差相同时，簇间距越小，“应力阴影”效应越强，水力裂缝扩展偏离主应力方向，裂缝宽度减小，不利于加砂；③天然裂缝内摩擦系数、天然裂缝与水力裂缝夹角越小，净压力越高，天然裂缝越容易开启或剪切，增加了改造裂缝的复杂性；④裂缝扩展方向、裂缝长度二者预测具有较好的一致性，但受天然裂缝发育差异的影响，局部压裂段有一定的差异性。所提出的裂缝预测方法为页岩改造裂缝识别与后期压裂设计提供了技术支撑。

Methodology of hydraulic fracture network prediction and its application

In shale reservoirs, natural fractures are developed, which, together with a complex fracture network induced by hydraulic fracturing, can provide effective pathways for oil and gas flowing. Due to the complex fracture network propagation, however, hydraulic fracture identification and fracturing design optimization are faced with new technical challenges. In this paper, a fracture network propagation model (FNPM) was established according to hydromechanics and fracture mechanics theory. In this model, the interaction between hydraulic fractures and natural fractures and the effect of stress shadow between fractures are considered comprehensively. By virtue of this model, fracture propagation direction, fracture network geometry, and propped fracture area in different conditions (e.g. mechanic parameters and natural fractures) can be predicted. Based on theoretical simulation and practical application, some conclusions are reached. First, stimulated fracture network propagation is controlled by multiple factors. For example, ground stress difference, cluster spacing and net pressure determine the stress shallow effect, and the interaction mechanism between natural fractures and hydraulic fractures decides the complexity of induced fracture networks. Second, when the ground stress difference is identical, the smaller the cluster spacing is, the stronger the stress shadow effect. As a result, hydraulic fracture propagation is deviated from the main stress direction, and fracture width is decreased, which are unfavorable for sand pumping. Third, the lower the internal friction coefficient of natural fractures and the included angle between natural fractures and hydraulic fractures are, the higher the net pressure, and the more easily the natural fractures can be opened or sheared, consequently the more complex the stimulated fractures are. And fourth, the predicted fracture propagation direction and fracture length are in good agreement, but they are locally different in some fracturing sections due to the development difference of natural fractures. The fracture prediction method proposed in this paper provides a technical support for induced fracture identification and later fracturing design of shale reservoirs.