The formation of hydraulic fracture-network is an important goal for stable and increased production of unconventional oil and gas resources. However, due to the extremely complex mechanical behavior of hydraulic-fracture propagation, there is still a lack of fine, effective and systematic evaluation methods on 3D fracture-network structure. The quantitative characterization of hydraulic-fractures is of great significance to reveal the law of fracture propagation and to evaluate the reservoir stimulation effectiveness reasonably. Although the micro-cracks evolution process follows a random rather than deterministic pattern, the overall fracture morphology is featuring statistical regularity. In this study, after the indoor hydraulic-fracturing simulation experiments of the tight sandstone specimens, 3D optical scanning was used to visualize the fracture-network. On this basis, two aspects of quantitative statistics of 3D complex fracture-networks were established by plane and sphere cuttings for the first time, including angle distribution statistics and spatial variation statistics. Comparative analyses with reported methods showed that, the overall deflection of the macroscopic fracture-networks varies synchronously with the local roughness of the main hydraulic fractures closest to the wellbore, which satisfied exponential function or power-law function relationship between them. The variation of the area proportion with the propagation radius reflects the specific fracture propagation process such as crack initiation, bifurcation and arrest, and could be used to assess the complexity of the fracture-network. Moreover, a new stimulation evaluation index E_s was established by comprehensively considering the stimulated-reservoir-area, the dispersion degree of the angle distribution and the complexity of spatial variation for 3D fracture-network. The insights gained warrant further applicability on hydraulic-fracturing in the actual engineering scale.

形成水力裂缝网络是非常规油气资源稳产增产的重要目标。然而，由于水力裂缝扩展的力学行为极其复杂，目前还缺乏精细、有效和系统的三维裂缝网络结构评价方法。水力裂缝的定量表征对于揭示裂缝扩展规律、合理评价储层改造效果具有重要意义。虽然微裂纹演化过程遵循随机而非确定的模式，但整体裂缝形态具有统计规律性。本研究经过致密砂岩标本，3D 光学扫描用于可视化裂缝网络。在此基础上，首次通过平面和球体切割建立了3D复杂缝网定量统计两个方面，包括角度分布统计和空间变化统计。与已有方法的对比分析表明，宏观裂缝网的整体挠度与最接近井筒的主要水力裂缝的局部粗糙度同步变化，它们之间满足指数函数或幂律函数关系。面积比例随扩展半径的变化反映了裂缝萌生、分叉、止裂等特定的裂缝扩展过程，可用于评估裂缝网络的复杂性。而且，E _s是综合考虑3D裂缝网改造的储层面积、角度分布的离散程度和空间变化的复杂性而建立的。获得的见解保证了水力压裂在实际工程规模上的进一步适用性。