Characterization of Multi-Scale Fractures in Deep Shale Reservoirs in North Area of Luzhou

天然裂缝作为页岩储层的关键储集空间与渗流通道，对油气运移、富集及开发方案具有决定性影响。四川盆地南部泸州北区块深层页岩储层受复杂地质作用影响，呈现多成因机制与显著尺度差异的裂缝发育特征，传统单一建模方法难以实现精准表征。本研究基于岩心分析、成像测井解释及三维地震数据，建立了多尺度离散裂缝网络建模技术体系，系统揭示了该区裂缝发育规律。主要成果包括：① 研究区裂缝系统遵循幂律分布规律，按规模可划分为大尺度与中小尺度两类；② 成因机制研究表明，中小尺度裂缝可划分为构造成因(占比近90%)与沉积成因两类，其中构造成因裂缝可进一步细分为褶皱控制型与断裂控制型；③ 创新性提出多属性体约束下随机–确定性耦合建模方法，通过引入神经网络模型有效降低裂缝预测的不确定性；④ 模型验证表明，研究区以中小尺度裂缝为主导，构造运动是裂缝发育的主控因素。As the key reservoir space and seepage channel of shale reservoir, natural fractures have a decisive influence on oil and gas migration, enrichment and development plan. The deep shale reservoir in Luzhou North Block, southern Sichuan Basin, is affected by complex geological processes and presents the characteristics of fracture development with multiple genetic mechanisms and significant scale differences, which makes it difficult to be accurately characterized by the traditional single modeling method. Based on core analysis, imaging logging interpretation and 3D seismic data, a multi-scale discrete fracture network modeling system is established to systematically reveal the fracture development law in this area. The main results include: ① The fracture system in the study area follows the power law distribution law, and can be divided into large scale and small scale according to the scale; ② The study of genetic mechanism shows that the small and medium-scale fractures can be divided into two types: tectonic origin (accounting for nearly 90%) and sedimentary origin, and the tectonic origin fractures can be further subdivided into folding controlled type and fault controlled type. ③ An innovative random-deterministic coupling modeling method under multi-attribute body constraints is proposed to effectively reduce the uncertainty of crack prediction by introducing a neural network model; ④ Model verification shows that the study area is dominated by small and medium scale fractures, and tectonic movement is the main factor of fracture development.