大66井区致密砂岩气藏测井解释模型研究

发布时间：2019-05-24 13:35

【摘要】：大牛地气田位于鄂尔多斯盆地,属于典型的低孔、低渗致密砂岩气藏,其勘探及开发越来越受到重视,致密砂岩气藏储层参数的测井解释模型建立一直是测井解释的难点。由于致密砂岩气藏与常规砂岩相比有很大的不同,孔隙度和渗透率相对较低,孔隙结构复杂,储层的储集性能也相对较差,这些特点都会给致密含气砂岩储层的测井解释方法研究带来困难。本文首先进行岩心归位,充分利用已有的测井资料、岩心资料、物性等资料,在对储层的岩石学、物性、电性特征进行深入研究的基础上,针对研究区致密砂岩储层低孔、低渗的特征,系统的对储层四性关系进行了研究,明确了储层物性的主要控制因素,为测井解释模型的建立打下了基础。为了更加准确的求取储层参数,本文采用分层位、分岩性来建立储层参数测井解释模型,对于砂岩类型的划分,采用了神经网络判别来进行岩性识别,取得了较好的分类效果。对于储层参数的求取,通过使用岩心资料采用了多种统计方法建立了物性参数模型,包括线性一元、多元回归以及非线性的BP神经网络和支持向量机。通过对多种模型的效果对比,选出适合研究区的测井解释方法,最后对解释模型效果进行验证,证明能够挖掘致密砂岩储层电性和物性间非线性关系的支持向量机方法预测储层物性参数有较好的预测效果,提高了大牛地致密砂岩气藏储层参数测井解释模型的精度。
[Abstract]:Daniudi gas field is located in Ordos Basin, which belongs to typical low porosity and low permeability tight sandstone gas reservoirs. More and more attention has been paid to its exploration and development. The establishment of logging interpretation model of tight sandstone gas reservoir parameters has always been a difficult point in logging interpretation. Because the tight sandstone gas reservoir is very different from the conventional sandstone, the porosity and permeability are relatively low, the pore structure is complex, and the reservoir performance of the reservoir is relatively poor. These characteristics will bring difficulties to the study of logging interpretation methods for tight gas-bearing sandstone reservoirs. First of all, this paper makes full use of the existing logging data, core data, physical properties and other data, on the basis of in-depth study of the petrology, physical properties and electrical characteristics of the reservoir, aiming at the low porosity of tight sandstone reservoir in the study area. The characteristics of low permeability and the systematic study of the four characteristics of reservoir are carried out, and the main controlling factors of reservoir physical properties are clarified, which lays a foundation for the establishment of logging interpretation model. In order to obtain reservoir parameters more accurately, this paper uses stratification and lithology to establish logging interpretation model of reservoir parameters. For the classification of sandstone types, neural network discrimination is used to identify lithology, and good classification effect is obtained. For the calculation of reservoir parameters, various statistical methods are used to establish physical parameter models, including linear univariate, multiple regression, nonlinear BP neural network and support vector machine. Through the comparison of the effects of various models, the logging interpretation method suitable for the study area is selected, and finally the effect of the interpretation model is verified. It is proved that the support vector machine method, which can excavate the nonlinear relationship between electrical property and physical properties of tight sandstone reservoir, has a good prediction effect on reservoir physical parameters, and improves the accuracy of logging interpretation model of reservoir parameters in Daniudi tight sandstone gas reservoir.
【学位授予单位】：中国石油大学(华东)
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P618.13;P631.81

【参考文献】