裂缝性储层属性分析与随机模拟

PROPERTY ANALYSIS AND STOCHASTIC MODELLING OF FRACTURED RESERVOIRS

与常规储层相比,裂缝的存在致使裂缝性储层非均质性增强,如何全面、准确地定量描述裂缝网络成为细致刻画这类储集层的关键。合理整合相关裂缝信息数据,运用最优概率分布函数对裂缝属性特征进行统计和空间分析,结合区域地质背景和研究目的,选择最佳理论模型是裂缝性储层建模的基本思想。依据这一原则对裂缝性储层随机模拟的方法进行了综合分析,认为:(1)非线性分析方法能够最大限度减小不同数据源、不同尺度数据整合所产生的误差,综合分析的裂缝数据较为理想;(2)运用圆周统计、人工智能分析裂缝属性能够获得较为合理的结论;(3)复合模型既能表现裂缝离散性特征,又能体现裂缝聚类的连续性效果,能较好地表征裂缝性储层特点。但裂缝发育的复杂性决定了裂缝储层建模还需综合不同学科、运用多种方法和手段进行深入的探讨和研究。

Compared with conventional reservoirs,fractured reservoirs have a higher heterogeneity due to the existence of fractures.The key of describing this type of reservoir lies in a complete and accurate quantitative description of fracture networks.The purpose of modeling fractured reservoir is integrating the information and data of relevant fractures,making a statistical analysis and spacial analysis of the fracture characteristics with the optimal probability distribution function,and choosing a theoretical model according to the regional geologic background and research purposes.According to this principle,a multidisciplinary analysis on the probabilistic simulation method of fractured reservoir is made.From the analysis,some conclusions are made as follows.First,the nonlinear analysis method can reduce the errors caused by the integration of multi-source and multi-scale data maximally.The fracture data analyzed comprehensively is comparatively good.Second,we can draw a relatively reasonable conclusion with circular statistics method and artificial intelligence method when we analyse fracture attributes.Third,compound model cannot only show the discreteness of fractures but also reflect the continuous effect of fractures.However,due to the complexity of the fracture development,fracture modeling still need to be improved by multi-disciplinary analysis and multi technique research.