统计岩石物理技术在薄储层定量解释中应用

Application of statistical rock-physics to quantitative interpretation in thin reservoir

储层物性参数(孔隙度、含流体饱和度等)的空间分布,直接影响油气储层的品质,是地球物理工作者进行储层评价的主要依据。由于储层物性参数和地震属性之间的关系是复杂的、非线性的,因此储层物性参数预测是当前研究难点。首先以地震岩石物理分析为基础,建立储层弹性参数与物性参数的先验概率密度函数,利用地质统计学反演得到高分辨率弹性参数体,在此基础上,利用贝叶斯分类算法,实现储层岩相及物性参数的定量预测,同时给出不确定性分析。通过薄储层实际工区应用,反演的岩性、物性数据体纵向特征与测井资料吻合较好,空间展布特征与该区的地质规律一致,不仅解决了薄储层的预测精度问题,而且实现了薄储层定量预测。

The reservoir property （porosity, fluid saturation, etc. ） is the most important parameter in evaluating the quali-ty of the oil and gas. It is the basic element in reservoir evaluation for geophysicists. Because the relation between reservoir property and seismic attributes is complex and nonlinear, how to get the accurate reservoir property is a difficult problem. In this paper, based on seismic rock physics, we used the Markov Chain Monte Carlo to build the probability density function （PDF） , which is about reservoir property and elastic parameters. We, then, use the geostatistical inversion to get elastic pa-rameter which is high resolution. And, finally, under the prior information, we use the Bayes theory to get the reservoir facies and properties. It is a quantitative prediction, which gives us uncertainty analysis by probability. In actual thin bed study area, we use this method to calculate the reservoir facies and properties. The result is consistent to the log data in well location, and the slice attribute is agree to the geology. So this method not only solved the problem of resolution of thin bed but also realized the quantitative prediction.