StatModcMC反演在薄互层地震储层预测中的应用.docVIP

StatModcMC反演在薄互层地震储层预测中的应用 　　摘要： 地质统计学反演首先应用确定性反演方法得到波阻抗体，以了解储层的大致分布，并用于求取水平变差函数；然后从井点出发，井间遵从原始地震数据,通过随机模拟产生井间波阻抗，再将波阻抗转换成反射系数并与确定性反演方法求得的子波进行褶积产生合成地震道，该方法有效地综合了地质、测井和三维地震数据，反演结果是多个等概率的波阻抗数据体实现，符合输入数据的地质统计学特征并受地质模型的约束，具有测井数据的垂向分辨率高和地震数据的横向分辨率高的优势，可用于不确定性评价. 　　Abstract: In geostatistical inversion, first, wave impedance is obtained by using deterministic inversion methods in order to understand the general distribution of reservoirs and to get horizontal variogram; then, interwell wave impendence is produced by stochasticsimulation based on original seismic data; third, the interwell wave impendence is transformed into reflection factor, the compose seismic traces are produced by the convolution of the wave impendence with the wavelets obtained by the deterministic inversion methods, the proper matching of the compose seismic traces with original seismic traces is reached by iterating again and again. The geostatistical inversion method integrates geologic, logging and 3D seismic data, and it combines the higher vertical resolution of logging data with the higher horizontal resolution of seismic data. The inversion result accords with the geostatictical characteristics of the input data and is restrained by geologic model. The several implementations of the inversion result are used for the nondeterministic assessment of the reservoirs. The applied result of the inversion method in Fuyu Oilfield in Jilin Province shows that, the geostatistical inversion method caneffectively predict reservoirs, especially thinner reservoirs by integrating geologic, logging and 3D seismic data. 　　关键词： 地质统计学反演；薄层砂体；随机模拟 　　Key words: geostatistical inversion；thin reservoir；stochastic simulation 　　 中图分类号：P315 文献标识码：A 文章编号：1006-4311（2011）28-0326-02 　　0 引言 　　StatModcMC反演是地质统计学反演的一种实现方法。它以地震反演为初始模型，从井点出发，井间遵从原始地震数据即以地震数据为硬数据（hard data），建立定量的波阻抗三维地质模型，进行储层横向预测[1]。其特点在于综合了地震反演与储层随机建模的优势，充分利用地震数据横向密集的特点，精确求取不同方向上的变差函数。 　　1 地质统计学反演 　　地质统计学是数学地质领域中一门发展迅速且有着广泛应用前景的新兴学科。随着对近地表速度模型反演精度要求的不