基于EEMD的高分辨率层序地层划分方法

划分不同级次的层序界面以及识别其内部的沉积旋回类型是层序地层分析的重要基础。测井信号是不同周期的地层旋回信息的叠合，通过时频分析方法可将测井数据内隐藏的地层旋回性信息拓展到不同的频带内，进而进行不同级次层序界面的划分以及旋回类型的判别。利用总体经验模态分解（Ensemble Empirical Mode Decomposition，EEMD）方法对测井数据进行分解，得到不同频带的本征模态分量，结合Hilbert变换获取各分量的频率特征。最后根据各分量的周期性波动和频率变化特征与不同沉积旋回类型以及各级层序界面之间的对应关系划分地层层序。在东营凹陷沙三段的应用中表明，该方法适用于短期、中期、长期基准面旋回的划分，为高分辨率层序地层划分提供了新思路。

Method for high-resolution sequence stratigraphy division based on Ensemble Empirical Mode Decomposition

Dividing sequence boundary of different levels and identifying internal types of sedimentary cycle are important bases of sequence stratigraphic analysis. Logging signals are the superposition of sedimentary cycles in different periods. It can be applied to classify the sequence boundaries of different levels,and identify the types of cycles by time-frequency analysis which can expand the information of stratigraphic cycle hidden in the well logging data into different frequency bands. Ensemble Empirical Mode Decomposition(EEMD) was applied to decompose well logging data into Intrinsic Mode Functions of different frequency banks,then the frequency information of each components can be obtained by Hilbert transform. Stratigraphic sequence was divided according to the relationship between the periodic fluctuation of each components and the different types of sedimentary cycle, as well as the relationship between the frequency characteristics of each components and sequence boundaries of all levels. The application in the third member of Shahejie Formation of Dongying Sag shows that this method is suitable for the cycle division of short-term,medium-term and long-term. It provides a new idea for division of high-resolution sequence stratigraphy.