【作者】 眭素文；

【导师】 张世红；

【作者基本信息】 中国地质大学（北京） ， 固体地球物理学， 2009， 博士

【摘要】 以地球物理测井曲线为主要研究对象,以识别地层中的米兰柯维奇旋回为主要目的,本文系统论述了目前使用比较多,效果比较明显的周期识别方法。通过对这些方法的理论依据、使用条件及计算效果进行研究,将其中分析效果明显、易于计算机实现的方法,通过改写公式,编制成各自独立的功能分析模块;以接口调用的方式将各功能模块集成到统一的Windows界面中。研制成一套界面友好、功能比较齐全的旋回识别软件系统。文中介绍了米兰柯维奇旋回的理论及研究背景,围绕旋回识别问题提出了三类旋回分析方法。系统讨论了经典的功率谱分析方法,针对其原理、应用前提以及方法不足进行了深入的分析研究。针对经典功率谱方法在频率域不能反映频谱随时间变化的缺点,讨论了能同时反映时频特征的短时傅立叶变化的理论和方法,同时指出了方法的不足。针对前面两类方法的缺点,研究了小波变换识别旋回的优势,重点讨论了Morlet小波变换的原理和方法。详细研究了对非等间距取样数据直接进行谱分析的方法。本文还对旋回分析计算中必要的预处理方法和窗函数选择方法进行了介绍。本文论述了地球物理测井数据作为旋回地层分析的依据和优势。在讨论了如何利用测井曲线识别沉积地层中米兰柯维奇旋回的前提下,首先研究了米兰柯维奇旋回与沉积速率的关系。详细讨论了如何从米兰柯维奇旋回的时频分布特点分析沉积速率随时间(深度)变化的特征。指出了利用沉积速率随时间(深度)变化的曲线将厚度剖面改正到时间剖面的理论依据。用不同构造环境下多口井的自然伽玛地球物理测井数据对松辽盆地青山口组和嫩江组一、二段进行了米兰柯维奇旋回分析,发现了地层中普遍记录了米兰柯维奇旋回。利用旋回分析结果分析计算了沉积速率变化特征;提取特定周期的自然伽玛测井曲线,进行高分辨率地层划分与对比。利用精度高、连续性好、较少受后期破坏的松科1井南井的自然伽玛测井数据分析计算了青山口组地层沉积时间,通过与已知资料对比,证明了计算结果的正确性;用岁差周期曲线高精度计算了青山口一段缺氧事件层段的持续时间,为此事件层段可能与白垩纪古海洋赛诺曼阶-土伦阶期界线附近的缺氧事件相对应提供了辅助证据。更多还原

【Abstract】 In this paper, the author gives the cycle identification method for detecting the Milancovitch cycles from geophysical logging data. After a detailed discussion of the theory, using condition and using effect of the methods, the author develops a friendly interface and multiple functional cycle identification program system by rewriting formulas, programming independent modules.In the beginning of the paper, the author discuses the theory background of the Milancovitch, and puts forward three kinds of period identification methods based on the problem of cycle identification. The author systematically discusses the theories, using conditions and deficiency of the classical power spectrum analysis methods. Then, according to the advantages of the classical power spectrum analysis methods, which there are no spectrum information changed with time in the frequency domain, the author studies the characteristics, and the advantages and disadvantages of short time fourier transform, a time-frequency analysis method. The author discusses the wavelet transform analysis method which overcoming the disadvantages of the two above methods. In the paper, the author also gives the discussion of preprocessing methods, window-function selection methods and power spectrum analysis methods for unequally-spaced data.The author discusses the evidence and advantage of geophysical logging data. As the precondition of geological application, the author discusses firstly, how to identify the Milankovitch cycles from the geophysical logging data, then, studies the relationship between Milankovitch cycles and deposition rate. The author introduces in detail how to analysis the characteristics of deposition rate changed with time(or depth)and puts forward to the thought of transforming the depth profile to time profile by use of the deposition rate curves changed with time.In the paper, the author calculates the Milankovitch cycles of Qingshankou formation and the first part to second part of Nenjiang formation in Songliao basin by logging data of different wells. The result shows the Milankovitch cycles are recorded in all the wells. Using the Milankovitch cycles result, the author analyses the feature of deposition rate changed with time, extracts the specific cycle curves for stratigraphic division and comparison with high precision. The author also explores the method of calculating the continuous time of sedimentary strata, and calculates the continuous time of Qingshankou formation by mean of the logging data from Songke NO.1 well with good continuity and high precision. Comparing to the known information proves the correctness of the result. By counting the number of wavelength of the precession cycles, the author analyses the continuous time of event key bed from the first part of Qingshankou formation, which provides supported evidence that this event key bed is corresponding to the paleo-oceanic anoxic event near the boundary of Cenomanian-Turonian in Cretaceous.更多还原