【作者】 赵德庆；

【导师】 王绪本； 胡平；

【作者基本信息】 成都理工大学 ， 地球探测与信息技术， 2010， 硕士

【摘要】 目前,从地震道提取属性参数是地震资料处理中的一种主要的技术方法。地震道从本质上讲是一个典型的非线性时间序列,用非线性的技术方法进行地震道属性参数提取才会对地震信息的合理利用更有利。本文是在非线性理论的基础上对非线性地震属性参数关联维的研究。首先对动力学非线性系统的混沌、分形特征以及地震信号非线性特征等理论进行了分析。表明地震信号是混沌的时间序列,具有混沌与分形的特征。其次在分析了地震属性,包括属性的分类,属性的提取之后,利用非线性的理论对地震数据进行非线性地震属性提取。以分形技术为基础,文章主要研究了地震道的分维数关联维的求取原理与过程。由于地震道是一个时间序列,一般时间序列主要是在时间域或变换域中进行研究,而在混沌时间序列处理中,各种不变量的计算都是在相空间中进行,因此相空间重构是混沌时间序列处理中非常重要的第一步。所以文章对相空间重构理论以及对相空间重构的参数的计算方法进行了分析。为分形维数关联维的求解提供了理论基础。在对实际的地震资料处理求取关联维时,一方面针对地震数据本身的特点,文中分析确定了计算中各个参数的取值情况。另一方面由于地震数据量非常大,利用传统的计算方法,时间太久。基于这个问题,分析了在误差允许的范围内,采用人机结合的方式以及对关联积分的计算进行改进以求取地震道的关联维数。在算法中我们采取了加窗与不加窗两种计算方式。文中首先运用该方法针对不同的理论模型进行数值试验。目的是分析这个非线性地震属性参数关联维在不同地质状况的响应情况。结果表明参数关联维能明显地反映地质状况。其次文章分析了利用此改进后的算法对实际地震资料求解关联维这个地震属性参数。求取的关联维结果能较好地与测井、地质资料吻合。所以说在地震勘探中,关联维数的研究对地震资料的处理与解释提供了较好的判断依据。更多还原

【Abstract】 At present, the seismic attribute extraction from the seismic trace is a main technical method in seismic data processing. The seismic trace essentially is a typical nonlinear time series. We use the seismic information reasonably by using the nonlinear method to extract the seismic attribute.We study the nonlinear seismic attribute correlation dimension based on the theory of nonlinear in this paper. And we first analyze chaotic characteristics and fractal characteristics of the dynamic Nonlinear System and the nonlinear characteristics of seismic signals. It Show that the seismic signal is chaotic time series with chaotic and fractal characteristics. Secondly we extract non-linear seismic attributes from seismic data by using non-linear theory after the analysis of seismic attributes, including the classification of seismic attributes and extraction of seismic attributes. This paper mainly studied the calculation principle and calculation process of seismic correlation dimension based on Fractal Technology. Phase space reconstruction is a very important first step in dealing with the chaotic time series. Because seismic is a time series and the general time-series mostly are studied in the time domain or transform domain; but all non-variable must be counted in phase space in chaotic time series processing. We analyze phase space reconstruction theory and the calculation of the phase space reconstruction parameters in the article. It provides a theoretical basis for solving the correlation dimension. On the one hand we analyze to determine the parameters in the calculation because of the characteristics of the earthquake data itself. On the other hand because of seismic data is very large it takes a long time for using the traditional method in the calculation. When we calculate the correlation dimension in the actual seismic data processing. For this problem we make use of man-machine combination within the permitted error and improve the calculation of correlation integral to obtain the correlation dimension of seismic traces. We adopt the calculation with window and the calculation without window in the algorithm.In order to analyze the nonlinear seismic attribute correlation dimension response to different geological conditions, we first conduct numerical experiments using this method on different models in the paper. And The results show that the correlation dimension can clearly reflect the geological conditions. Second, we use the improved algorithm to calculate the seismic attribute correlation dimension in the actual seismic data processing. And the correlation dimension results consistent with the logging data and geological data. Therefore, the research of the correlation dimension can provide a better basis to judge in the processing and interpretation of seismic data in seismic exploration.更多还原