离散裂缝网络随机建模方法

Discrete Fracture Network Stochastic Modeling

离散裂缝网络模型可以表征裂缝性油藏任意尺度上的非均质性。在前人工作的基础上,对离散裂缝网络随机建模方法做了进一步的研究:考虑了局部断层(或大裂缝)等不连续面对裂缝分布的影响,采用空间变概率的方法确定裂缝位置;采用循环统计方法分析裂缝产状数据。针对裂缝不同属性的特征,分别采用了Fisher分布模型和分形的方法对裂缝的产状和大小进行模拟。计算实例显示,空间变概率方法确定裂缝位置能较好地表征局部断层(大裂缝)对裂缝分布的影响;循环统计方法能有效地避免建模过程中线性统计造成的误差。

Discrete Fracture Network (DFN) Model can be used to describe reservoir heterogeneity at any scale.On the basis of the previous work,further study on DFN stochastic modeling was made in this paper.Space Variable Probability Method was used to account for the effects of the local faults (large fractures) on distribution of fractures.Circular Statistics was applied for fracture orientation data analysis.According to different characteristics,fracture orientation and size was modeled based on Fisher distribution and fractal dimension respectively.It was showed that Space Variable Probability Method was more reasonable in simulating the effects of local faults (large fractures) on distribution of fractures,and statistical bias caused by linear statistics could be avoided effectively via Circular Statistics.