分形在油藏动态参数预测中的应用

Application of Fractal Character in Prediction of Reservoir Operating Parameters

由于储层非均质性具有分形特征 ,应用分形理论描述储层非均质性是最有效的工具之一。采用频谱分析将储层参数(孔隙度、渗透率、含油饱和度等 )进行分析 ,计算出 β和分形特征指数 H进行分形模型的识别。分形模型包括分数布朗(Brown)运动 (Fbm)和分数高斯 (Gauss)噪声 (Fgn)。其识别的定量标准是 ,当 1≤ β≤ 3时 ,数据序列为 Fbm;当 - 1≤ β≤1时 ,数据序列为 Fgn。识别出分形模型后 ,应用 R/S分析、方差分析和频谱分析求取 3个 H值 ,取其平均值以提高准确度。利用分形插值理论对储层参数进行分形插值模拟 ,得到储层参数的横向分布。

Describing heterogeneity of reservoir by fractal theory is one of the most effective methods because heterogeneity of reservoir has fractal nature. β and fractal nature index H can be figured out according to spectrum analysis of reservoir parameters (porosity,permeability, oil saturation, etc.). Fractal pattern includes Fraction Brown Movement (Fbm )and Fraction Gauss Noise( Fgn). The quantification standards to identify fractal pattern is: while 1≤ β ≤3, sequential data is Fbm; while -1≤ β ≤1, sequential data is Agn. After identifying fractal pattern, three H values can be acquired through R/S analysis, variance analysis and spectrum analysis, thereby, average value of which can be used to improve calculation precision. Fractal interpolation theory is used to simulate reservoir so to obtain transverse distribution of reservoir parameters.