苏里格致密砂岩储层BFA-CM混合最优化测井解释方法研究

发布时间：2018-05-20 09:30

本文选题：致密砂岩 + 细菌觅食算法　；参考：《吉林大学》2015年硕士论文

【摘要】：随着大型油气田勘探开发程度的不断深入，常规油气资源储量日益减少，人们对非常规油气资源的关注越来越多。致密砂岩气是一种典型的非常规油气资源，资源储量巨大，具有广阔的发展前景。致密砂岩气藏具有非均质性强、低孔低渗、孔隙结构复杂且多种泥质分布形式共存等特点，这些特点造成其测井评价困难，储层参数计算准确率不高。最优化测井解释方法是评价致密砂岩储层的一种有效途径。与利用有限测井曲线信息的传统顺序式测井解释方法不同，最优化测井解释方法依据地球物理反演理论并综合利用多种测井信息、地质资料和工作经验，运用最优化方法计算储层参数。最优化测井解释方法对测井信息的利用率高，可以灵活地变化解释模型和解释方法。它还可以进行自我质量检测，在生产实践中展示了独特的优势，并得到了广泛的应用。本文以苏里格致密砂岩储层盒8及山1层段为研究对象。整理并分析研究区内的测井数据、压汞和相渗实验数据、岩心物性分析以及试气结论等资料，确定致密砂岩储层含气特征及有效储层下限值，对有效储层进行基于结构约束的BFA-CM混合最优化测井解释方法评价。泥质的分布形式对储层参数具有重要影响，因此可以将其考虑到储层测井解释模型中以提高储层未知参数的计算精度。依据所建立的测井解释模型，可以得到中子、密度和声波的测井响应方程。在求解过程中需要对未知参数进行限制，以便求取得到的结果具有一定的合理性。除基本的数学物理约束条件外，将Thomas-Stieber结构推导得到的两个响应方程作为结构约束条件引入到最优化测井解释方法中，同时解决体积分数和岩石泥质结构。这样，成分体积和泥质结构分析可以在一个步骤中进行，实现了更全面更合理的岩石物理解释。建立最优化测井解释方法的数学模型后，需要选择合适的最优化方法求解数学模型的最优解。细菌觅食算法(BFA)是新兴的一种仿生类优化算法，通过模拟细菌在生物体内的生存过程来迭代寻求最优解，还未被应用于最优化测井解释方法中。实践表明，细菌觅食算法在寻优后期收敛速度变慢。为提高计算精度和效率，，本文将其与具有极强局部搜索能力的复合形算法(CM)相结合构成BFA-CM混合算法，作为最优化测井解释方法中的寻优方法。应用结合结构约束条件的BFA-CM混合最优化测井解释方法对苏里格致密砂岩储层实际资料进行处理。与没有结构约束条件的BFA-CM混合最优化测井解释方法相比，结合结构约束条件的处理结果更加稳定，与岩心分析及薄片数据吻合程度更好。
[Abstract]:With the deepening of exploration and development of large oil and gas fields, the reserves of conventional oil and gas resources are decreasing day by day, and people pay more and more attention to unconventional oil and gas resources. Tight sandstone gas is a typical unconventional oil and gas resource with huge reserves and broad development prospects. The tight sandstone gas reservoir is characterized by strong heterogeneity, low porosity and low permeability, complex pore structure and coexistence of various shaly distribution forms. These characteristics make the logging evaluation difficult and the accuracy of reservoir parameter calculation not high. Optimization logging interpretation method is an effective way to evaluate tight sandstone reservoir. Different from the traditional sequential logging interpretation method using finite log curve information, the optimal logging interpretation method is based on geophysical inversion theory and synthetically utilizes various logging information, geological data and working experience. The reservoir parameters are calculated by optimization method. The optimal logging interpretation method has a high utilization rate of logging information and can flexibly change the interpretation model and interpretation method. It can also be used for self-quality testing, showing unique advantages in production practice, and has been widely used. In this paper, Sulige tight sandstone reservoir box 8 and Shan 1 are taken as research objects. Well logging data, experimental data of mercury injection and phase permeability, core physical properties analysis and gas test conclusion were collected and analyzed to determine the gas-bearing characteristics of tight sandstone reservoir and the lower limit value of effective reservoir. The effective reservoir is evaluated by BFA-CM mixed optimization logging interpretation method based on structural constraints. The distribution of mudstone has an important influence on reservoir parameters, so it can be taken into account in the reservoir log interpretation model to improve the calculation accuracy of unknown reservoir parameters. The log response equations of neutron density and acoustic wave can be obtained according to the log interpretation model. In the process of solution, the unknown parameters should be restricted so that the obtained results are reasonable. In addition to the basic mathematical and physical constraints, the two response equations derived from the Thomas-Stieber structure are introduced into the optimal logging interpretation method as structural constraints, and the volume fraction and the muddy structure of the rock are solved at the same time. In this way, the analysis of composition volume and muddy structure can be carried out in one step, and a more comprehensive and reasonable interpretation of rock physics can be achieved. After establishing the mathematical model of the optimal logging interpretation method, it is necessary to select a suitable optimization method to solve the optimal solution of the mathematical model. Bacterial foraging algorithm (BFA) is a new kind of bionic optimization algorithm. It can find the optimal solution iteratively by simulating the survival process of bacteria in organism. It has not been applied to the optimal logging interpretation method. The practice shows that the convergence rate of bacterial foraging algorithm slows down in the later stage of optimization. In order to improve the calculation accuracy and efficiency, this paper combines it with the complex algorithm with strong local search ability to form the BFA-CM hybrid algorithm, which is used as the optimization method in the optimization logging interpretation method. The practical data of Sulige tight sandstone reservoir are processed by BFA-CM mixed optimization logging interpretation method combined with structural constraints. Compared with the BFA-CM mixed optimization logging interpretation method without structural constraints, the processing results with structural constraints are more stable, and are in better agreement with core analysis and sheet data.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P618.13;P631.81

【参考文献】