二维核磁共振岩心流体评价方法研究

发布时间：2018-01-13 04:11

本文关键词：二维核磁共振岩心流体评价方法研究　出处：《东北石油大学》2015年硕士论文　论文类型：学位论文

【摘要】：核磁共振测井在测量上较常规测井受到岩石骨架影响更小,因此在预测储层物性参数方面更加准确。通常在缺少岩心资料的情况下,以核磁共振测井资料计算的储层物性参数作为参照。一维核磁共振根据孔隙流体的横向弛豫时间的不同,利用反演出的弛豫谱对地下储层中的各种孔隙比例、渗透率参数以及含油气情况等进行预测。但由于测量技术问题,一维核磁测井数据信噪比低,并且测量的流体在弛豫信号上常常有较多的重叠区域,以致应用像差谱、移谱等方法所给出的解释结果不够准确。二维核磁共振的提出很好地解决了一维核磁共振在流体识别与评价上的不足。但目前的二维核磁反演方法不够成熟,在反演速度和精度上还存在很多问题。针对这点,本文以经典的子空间类型方法中的TSVD方法为研究基础,通过深入分析其奇异值截断与迭代算法结构后,创造性的提出了奇异值截断、分组处理算法和变参量迭代算法。对新方法的反演性能检测进行了数值模拟和岩石实验两个环节。在数值模拟中,新方法能够快速地还原出不同维数的构造谱,反演出的流体信号的分布范围和峰值与构造谱基本吻合,说明反演结果可靠;在岩心实验方面,能够依据该方法反演出的二维谱算出较为准确地饱和度参数,其相对误差控制在5%以内。检测结果表明,新方法理论性正确,能够快速处理二维核磁共振岩心实验资料,其孔隙流体的识别与评价结论可靠度高,不仅体现出该方法具有一定的实际应用价值,还为其他二维核磁反演方法研究提供了思路。
[Abstract]:Nuclear magnetic resonance logging is less affected by rock skeleton than conventional logging, so it is more accurate in predicting reservoir physical parameters, usually in the absence of core data. The physical parameters of reservoir calculated by nuclear magnetic resonance (NMR) logging data are taken as reference. According to the different transverse relaxation time of pore fluid, one-dimensional nuclear magnetic resonance (1D NMR) is used to determine the various porosity ratios in underground reservoir by using the inverse relaxation spectrum. Permeability parameters and oil and gas conditions are predicted, but due to the technical problems of measurement, the signal-to-noise ratio of one-dimensional nuclear magnetic logging data is low, and the measured fluid often has more overlapping areas on relaxation signals. So that the aberration spectrum is applied. The results obtained by the methods such as spectral shift are not accurate enough. The two-dimensional nuclear magnetic resonance (T1NMR) is a good solution to the problem of fluid identification and evaluation. However, the current 2D NMR inversion method is not mature enough. There are still many problems in the speed and precision of inversion. In this paper, based on the classical subspace type TSVD method, the singular value truncation and iterative algorithm structure are deeply analyzed. The singular value truncation, grouping processing algorithm and variable parameter iterative algorithm are creatively proposed. The inversion performance detection of the new method is simulated by numerical simulation and rock experiment. The new method can quickly reduce the structural spectrum of different dimensions, and the distribution range and peak value of the inverse fluid signal are basically consistent with the structural spectrum, which shows that the inversion results are reliable. In core experiment, the saturation parameters can be calculated accurately according to the two-dimensional spectrum of the method, and the relative error is controlled within 5%. The test results show that the new method is correct in theory. It can process 2D nuclear magnetic resonance core experiment data quickly, and its pore fluid identification and evaluation results have high reliability, which not only shows that this method has certain practical application value. It also provides an idea for the study of other two-dimensional nuclear magnetic inversion methods.
【学位授予单位】：东北石油大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TE311

【相似文献】