陆地地震勘探随机噪声时空特性研究

发布时间：2018-02-24 11:33

本文关键词： 随机噪声高斯性平稳性功率谱密度相似性线性　出处：《吉林大学》2016年博士论文　论文类型：学位论文

【摘要】：陆地地震勘探是资源勘探的重要手段。面对复杂的地表和近地表地质条件、越来越深以及空间越来越复杂的勘探目标,给地震勘探提出更高的要求。在陆地地震勘探的整个过程(采集、处理、解释)中,地震资料处理是至关重要一环,特别是处理过程中随机噪声的压制效果直接影响着后续一系列处理,包括速度分析,偏移成像以及反演等。然而,野外采集地震数据时,检波器接收到的地震记录中含有大量随机噪声信息,特别是勘探目标比较复杂时,信号很弱而噪声很强,记录整体信噪比较低,如何在低信噪比条件下有效压制随机噪声对地震勘探而言就显得尤为重要。由于信噪比较低,信号能量较弱,为了能有效的压制随机噪声就需要对地震勘探随机噪声的时空特性有较系统的认识和研究,从而指导随机噪声压制方法选择、改进以及压制策略。目前,对于随机噪声特性还未进行过科学系统的研究,这直接影响地震记录信噪比的有效提高以及随机噪声消减技术的发展。有效的改进随机噪声消减技术是建立在对地震勘探随机噪声特性准确认识的基础上。因此,关于地震勘探随机噪声特性的研究具有重要的实际意义。本文主要针对陆地地震勘探随机噪声,对其时域、空间域以及时空域特性进行详细系统的研究。主要对随机噪声的高斯性、平稳性、线性特性、功率谱特性和相似性进行综合分析。对比不同采集环境(草原、沙漠、山地、林带)下,随机噪声时空分布特性的差异,分析采集环境与随机噪声时空特性的关系。研究中,分析采集环境复杂程度和噪声时长等因素对于随机噪声特性的影响。最后,结合随机噪声生成机制和采集环境的特点,对研究结果的合理性进行综合论证。具体研究成果如下:将高阶统计量同Shapiro-Wilk检验法相结合来判断地震勘探随机噪声的高斯性。首先对地震勘探随机噪声的峰度值和偏度值高阶统计量的特性进行分析,发现随机噪声在高阶统计量分布方面同高斯随机过程之间存在明显差异。在此基础上,应用Shapiro-Wilk检验法对随机噪声的高斯性进行准确的判断。结果表明,随机噪声的高斯性受到采集环境特点的影响。采集环境越简单,随机噪声的非高斯性越明显;采集环境越复杂,随机噪声的非高斯性越弱。通过对不同地区随机噪声记录的高斯性检验结果进行对比分析,发现非高斯随机噪声记录占比普遍很高。因此,陆地地震勘探随机噪声应该被认为是非高斯随机过程。本文首次将基于时频分析和替代数据的平稳性检验方法引入到陆地地震勘探随机噪声的平稳性判断中,并对不同时长随机噪声的平稳性进行判断。检验方法的基本思想是通过比较随机噪声序列同替代数据在时频特性上的差异,从而对随机噪声序列的平稳性进行判断分析。通过对不同地区随机噪声的平稳性进行统计分析,研究结果表明,陆地地震勘探随机噪声不是平稳随机过程,其特性随着随机噪声时长和采集环境特点发生变化。复杂环境中的随机噪声非平稳性较强,而简单环境中的随机噪声平稳性较好。同时,对短时长和长时长随机噪声的平稳性进行对比分析,结果显示,短时长随机噪声的平稳性要好于长时长随机噪声的平稳性。后续实验结果与陆地地震勘探随机噪声平稳性结论相一致。选用Thomson谱估计方法对陆地地震勘探随机噪声的功率谱特性进行研究。传统上,通常认为陆地地震勘探随机噪声在时空域是满足白噪特性的。研究结果表明,陆地地震勘探从功率谱特性上看并不是白噪过程。随机噪声的功率谱特性随着采集环境特点的不同而发生变化,对于沙漠地区、草原地区等简单的地理环境而言,陆地地震勘探随机噪声的能量主要集中在低频处,可以认为是低频、窄带色噪声;对于林带地区、山地地区等复杂的地理环境而言,随机噪声中高频成分相对丰富,频带较宽,可以近似认为是宽频色噪声。同时,对陆地地震勘探随机噪声的时空相似性进行科学分析。研究结果表明,陆地地震勘探随机噪声的相似性主要受采集环境特点的影响。对于简单采集环境而言,随机噪声表现出一定的相似性;而对于复杂采集环境而言,随机噪声的时空相似性较差。在此基础上,本文结合采集环境特点和随机噪声平稳性研究结论,对随机噪声相似性统计结果的合理性进行必要的论证。采用延时矢量方差法和量化延时矢量方差方法,对陆地地震勘探随机噪声的线性特性进行研究,并对不同地区随机噪声的线性特性进行综合判断。研究结果表明,随机噪声不能简单的归为线性或非线性随机过程。随机噪声的线性特性同采集环境的特点密切相关。当采集环境相对简单时,随机噪声的线性特性较好,可以近似认为是线性随机过程;当采集环境复杂时,随机噪声中有非线性成分存在,简单的线性模型已经无法表征出随机噪声的特性,这时随机噪声应该被认为是非线性随机过程的产物。本文对陆地地震勘探随机噪声的高斯性、平稳性、线性、功率谱特性和相似性进行科学系统的研究。在研究中,对不同典型地理环境下采集到的随机噪声的特性进行判断,同时对不同地区随机噪声在特性方面的差异进行对比分析,得到关于陆地地震勘探随机噪声特性的总体认识。研究结果表明,传统上对于随机噪声特性的认识有一定的片面性,存在一定的局限性。随机噪声的特性是同采集环境密切相关的,不同时长随机噪声在特性方面也存在一定的差异。在此基础上,通过一系列的理论分析和正演实验,论证本文研究结果的正确性和可靠性。综上,本文的研究成果是对陆地地震勘探随机噪声特性的一种新认识,是对传统理论的一种有益的补充,为改进随机噪声消减技术和随机噪声建模领域的研究提供理论依据,具有较好的实际意义和应用前景。
[Abstract]:Land seismic exploration is an important means of resource exploration. In the face of complex surface and near surface geological conditions, more and more deep space and increasingly complex exploration targets for seismic exploration and put forward higher requirements. In the whole process of land seismic exploration (acquisition, processing, interpretation), seismic data processing is the most important link in particular, the random noise in the process of suppression effect directly affects a series of processing, including velocity analysis, migration imaging and inversion. However, the field acquisition of seismic data, with a large number of random noise information detector received seismic records, especially the exploration target is complicated, and the noise signal is very weak the record is very strong, the overall low SNR, how in the condition of low SNR effectively suppress random noise of seismic exploration is very important. Because of low SNR signals The amount is weak, in order to effectively suppress random noise of seismic exploration requires the spatial and temporal characteristics of random noise recognition and a systematic study, so as to guide the random noise suppression methods, and improved suppression strategy. At present, the random noise characteristics has been done on the science of system, which directly affect the signal-to-noise of seismic records the ratio of effective improvement and development of random noise reduction technology. The improved random noise reduction technology effectively is based on the random noise characteristics of an accurate understanding of the seismic exploration. Therefore, study on seismic random noise characteristics has important practical significance. This paper focuses on the land seismic random noise, the time domain, research with the system in spatial domain and space-time domain. The main characteristics of the random noise of Gauss, stability, linear characteristics, spectrum characteristics and similarity Comprehensive analysis. Comparison of different acquisition environment (grassland, desert, mountain, forest), differences in the spatial and temporal distribution characteristics of random noise, analysis of the relationship between acquisition environment and random noise. The spatial and temporal characteristics of the study, analysis of the influence of acquisition environment complexity and noise factors such as long to random noise characteristics. Finally, combined with the characteristics of random noise generation mechanism and acquisition environment, the rationality of the results of the comprehensive discussion. The specific research results are as follows: high order statistics with the Shapiro-Wilk test method to determine the combination of Gauss seismic random noise. Firstly, seismic random noise value and kurtosis skewness characteristics of high order statistics analysis, found the random noise in high order statistic distribution of significant differences between the same Gauss stochastic process. On this basis, use the Shapiro-Wilk test for random noise The accurate judgment of Gauss Gauss. The results show that the random noise is affected by the acquisition environment characteristics. Acquisition environment is simple, non Gauss random noise is more obvious; acquisition environment more complex, non Gauss random noise is weak. Through the analysis and comparison of the results of Gauss test for different regions of random noise the record, Gauss found that non random noise recorded very high proportion of common land. Therefore, seismic random noise should be considered non Gauss random process. In this paper, for the first time based on the stationary test side frequency analysis and surrogate data method is introduced to determine the stability of the land seismic random noise, and judge of long stationary random noise. The basic idea of different test methods are compared with the data in the random noise sequence instead of time-frequency characteristics of the differences of random noise sequence We can analyze the stationarity. Through statistical analysis of stationary random noise in different regions, the results show that the land seismic random noise is a stationary stochastic process, the characteristics of random noise with long time and collecting environmental characteristics change. The random noise in complex environment nonstationarity and stationarity, simple random noise the environment is good. At the same time, the comparative analysis of stationary random noise, long short long and long time results show that stationary stationary random noise in a short time is long long of random noise is consistent with the experimental results. The subsequent land seismic random noise stationarity. Conclusion using Thomson spectrum estimation spectrum the characteristics of power of a terrestrial seismic random noise is studied. Traditionally, usually think land seismic random noise in time-space domain is to meet the characteristics of white noise The research results show that the power spectral characteristics of land seismic exploration from the point of view is not white noise. The random noise power spectrum characteristics of different features with the collection of environmental changes, the desert area, simple geographical environment of grassland areas such as land, seismic random noise energy is mainly concentrated in the low frequency, can that is the low frequency, narrow band noise; for forest area, mountainous area and complex geographical environment, relatively rich in high frequency random noise, wide frequency band, can be considered as the broadband noise. At the same time, the spatio-temporal land seismic random noise similarity was analyzed. The results show that similarity land seismic random noise is mainly affected by the acquisition environment characteristics. For the simple acquisition environment, random noise showed some similarities; and for the acquisition of complex Environment, time and space of random noise similarity is poor. On this basis, combining with the conclusion acquisition environment characteristics and random noise stability research, the rationality of the similarity of the statistical result of random noise. The necessary arguments by delay vector variance method and quantified delay vector variance method, to study the characteristics of linear random land seismic exploration the noise characteristics of different regions, and the linear random noise comprehensive judgment. The results show that the random noise can not be simply classified as linear or nonlinear stochastic process. Closely related to characteristics of linear characteristics of random noise with the acquisition environment. When the acquisition environment is relatively simple, good linear characteristics of random noise, can be regarded as the is a linear stochastic process; when the acquisition environment is complex, there are nonlinear random noise, a simple linear model has been unable to represent The characteristics of random noise, the random noise should be considered to be the result of nonlinear stochastic process. In this paper, Gauss, on the land of seismic random noise stationarity, linearity, power spectral characteristics and research of the scientific system of similarity. In the study, characteristics of the random noise of different typical physical environment to the acquisition of at the same time to judge, on different areas of random noise in the characteristics of the comparative analysis, have overall understanding about the land seismic random noise characteristics. The results show that the traditional understanding of the characteristics of random noise has certain one sidedness and limitations. The property of random noise is the same acquisition environment the same random noise also exist certain differences in characteristics. On this basis, through a series of theoretical analysis and experimental demonstration of this research forward. The correctness and reliability of the results. In conclusion, the research results of this paper is a kind of new understanding of land seismic random noise characteristics, is a useful supplement to the traditional theory, for the improvement of random noise and random noise reduction technology in the field of modeling provides a theoretical basis, practical significance and good application prospect.

【学位授予单位】：吉林大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：P631.4

【相似文献】