基于多点地质统计学的密井网储层建模方法研究

Research on Reservoir Modeling in Dense Well Pattern Area Based on Multi-Point Geostatistics

系统介绍了多点地质统计学的基本原理以及与传统建模方法的区别。以胜利油田A区为例,在利用训练图像指导模拟的基础上,深入分析了砂泥岩比例、参考比例、垂向比例函数、概率趋势体等各项敏感参数对模拟结果的不同影响,提出了通过调整敏感参数来提高预测结果准确性的新方法;并从训练图像选取、提高模型确定性、无井控制区随机模拟等方面开展研究,确定了适合研究区的最优参数。同时,利用平面展布特征分析、不同模拟方法结果对比、抽稀井验证等方法,对研究区沉积相模型的预测精度进行了验证。实践结果表明,利用该方法得到的沉积相模型能够实现砂体平面展布特征和纵向叠置关系的准确预测,预测结果与实钻井及实际地质认识基本吻合,同时对于河流相砂体“顶平底凸”的地质特征也具有较好表征,对于密井网区河流相砂体研究具有较大的指导意义,可为其他同类型油藏的储层预测研究工作提供方法指导。

This paper introduces the basic principle of multi-point geostatistics and its difference with traditional modeling methods. Taking Gudong 2 th block in Shengli Oilfield as an example, on the basis of using training images to guide the simulation, In this paper, the different influences of sensitive parameters such as sand-shale ratio, reference proportion, vertical proportion function and probability trend body on the simulation results are deeply analyzed, and a new method is proposed to improve the accuracy of the prediction results by adjusting sensitive parameters. The optimal parameters for the study area were determined by the selection of training images, the improvement of model certainty, and the stochastic simulation of the non-well controlled area. At the same time, the prediction accuracy of the sedimentary facies model in the study area was verified by the analysis of plane distribution characteristics, the comparison of the results of different simulation methods, and the verification of reduced wells.Practice results show that using this method to get the model of sedimentary facies, sand body can be implemented plane distribution characteristics and vertically superimposed relationship of accurate prediction. The predicted results are basically consistent with the actual drilling and geological understanding, and the geologic features of fluvial facies sandstone also has a good characterization. It is of great guiding significance to the study of fluvial facies sandstone in dense well pattern area and can provide methodological guidance for the research of reservoir prediction of other similar reservoirs.