序贯指示模拟在储层沉积相建模中的研究及应用

发布时间：2018-09-04 13:45

【摘要】：目前我国多数老油田已进入高含水开发阶段，勘探和开发剩余油是这一阶段的重点工作。随着油田井网的加密，利用加密井提供的地质信息，通过储层沉积相建模建立储层砂体模型，能够对分布在砂体中的剩余油的勘探和开发起到重要作用。结合油田项目，调研了国际上比较流行的建模软件——斯伦贝谢公司的Petrel软件、挪威Roxar公司的RMS地质建模软件，以及我国自主研发的建模软件——GPTModel油藏地质建模软件、数字油藏表征工具Direct，了解了这些建模软件中沉积相建模的一般步骤和常用算法。最终选用Petrel软件作为本次研究的建模工具。研究开始前首先收集了某油田区块建模的相关数据，包括井位坐标、补心海拔、井轨迹、测井曲线、地震解释断层数据、地震解释控制面数据等，，通过编程完成这些数据的格式转换，并修正了部分错误数据。然后应用GPTMap软件对测井曲线进行分层得到小层分层数据和断点数据。接着在Petrel中根据地震解释的断层数据应用Fault modeling模块建立断层模型；按照网格精度要求应用Pillar gridding模块建立三维构造模型；根据分层数据和断点数据以及地震解释的控制面数据应用Make horizons模块建立地层模型。最后采用确定性建模方法将GPTMap中的沉积相图导入Petrel中建立储层沉积相模型，同时还采用随机建模方法，具体是采用序贯指示模拟方法建立储层沉积相模型。重点研究了序贯指示模拟算法，对该算法以及该算法的地质统计学基础知识包括变差函数的计算、实现克立金估计等的数学原理进行了较为详细的说明，并使用MATLAB工具计算了实验变差函数，编程实现了普通克里金估计方法和序贯指示模拟方法。本次研究的主要成果是建立了两种沉积相模型。通过与确定性建模方法得到的模型进行比较，序贯指示模拟的模型与地质专家确定的沉积相模型是比较相符的。在数据量比较大的情况下，地质专家可依据序贯指示模拟的结果，手工修改沉积相图，节约大量时间提高工作效率。
[Abstract]:At present, most of the old oil fields in China have entered the stage of high water cut development, and the exploration and development of remaining oil is the key work in this stage. With the infilling of oilfield well pattern and geological information provided by infilled well, the reservoir sand body model can be established by reservoir sedimentary facies modeling, which can play an important role in the exploration and development of remaining oil distributed in sand body. Combined with the oil field project, the international popular modeling software-Schlumberger 's Petrel software, the Norwegian Roxar's RMS geological modeling software, and the independently developed GPTModel reservoir geological modeling software are investigated. The digital reservoir characterization tool Direct, understands the general steps and common algorithms of sedimentary facies modeling in these modeling software. Finally, Petrel software is chosen as the modeling tool of this study. Prior to the beginning of the study, the relevant data of modeling in a certain oilfield block were collected, including well location coordinates, supplementary altitude, well trajectory, logging curve, seismic interpretation fault data, seismic interpretation control surface data, etc. Through programming to complete the format of these data conversion, and fixed part of the error data. Then GPTMap software is used to stratify the log curve to obtain the data of small layer stratification and fracture point. Then the fault model is established by using Fault modeling module in Petrel according to the seismic interpretation fault data, and the 3D structural model is built by using Pillar gridding module according to the requirements of grid precision. The stratigraphic model is established by using Make horizons module based on stratified data, fracture data and control surface data of seismic interpretation. In the end, the sedimentary facies diagram in GPTMap is introduced into Petrel to establish reservoir sedimentary facies model, and the stochastic modeling method is used to establish reservoir sedimentary facies model. The sequential indicator simulation algorithm is studied in detail. The basic knowledge of geostatistics of the algorithm, including the calculation of variation function, and the mathematical principle of realizing Kriging estimation, etc., are explained in detail. The experimental variation function is calculated by using MATLAB tool, and the general Kriging estimation method and sequential indicator simulation method are realized by programming. The main results of this study are the establishment of two sedimentary facies models. Compared with the model obtained by deterministic modeling method, the sequential indicative simulation model is in good agreement with the sedimentary facies model determined by geological experts. In the case of large amount of data geological experts can manually modify the sedimentary facies diagram according to the results of sequential indicator simulation and save a lot of time to improve the working efficiency.
【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

【参考文献】