同步挤压小波变换及其在薄互层中的应用

发布时间：2018-06-28 11:59

本文选题：同步挤压小波变换 + 参数设置　；参考：《中国石油大学(华东)》2015年硕士论文

【摘要】：在地震信号分析和解释领域,小波变换一直都是人们研究的重点与热点,于是相继出现了许多小波变换类的时频分析方法,它们或是以小波为基础进行深入的改进,或是联合小波与其它方法的组合。从连续小波变换、三参数小波变换到S变换,甚至是广义S变换,都是以小波变换为基础而发展起来的,而且实际也证明了这些小波变换类的方法在对信号进行时频分析时效果越来越好,分辨率越来越高。但为了满足更高的要求,出现了一种联合小波变换和重组的新方法—同步挤压小波变换。本文主要从同步挤压小波变换的原理出发,研究了该算法所具有的相关性质,如稳定性和可逆性等,通过模型分析了解到重构误差非常小。同时也研究了算法中的相关参数对时频分析结果的影响,如小波母函数、小波母函数的参数选择和小波阈值等。通过分析讨论,了解到对地震信号进行分析时,一般选择Morlet小波;而中心频率越大,对瞬时频率的识别能力越强;同时,合适的小波阈值能有效地抑制噪音的干扰。因此,对这些参数的合理设置,将大大地增强同步挤压小波变换的应用效果。最后,针对同步挤压小波变换在分析瞬时频率的变化率不为零的信号时存在的缺陷,研究了广义同步挤压小波变换,但由于算法中的参数是根据待分析信号的特征来确定的,故其应用性不强。对模拟信号和实际信号分别采用连续小波变换、三参数小波变换、S变换和同步挤压小波变换进行时频分析,通过对比看出了同步挤压小波变换的优越性,即具有更高的时频分辨率。本文最后研究了薄层的振幅响应特征和频率响应特征,因这两者分别是从时间域和频率域来刻画薄层反射信号的特征,而时频分析技术正好同时利用了二者的性质,故将同步挤压小波变换应用于薄互层的瞬时谱分析中,通过对实际地震资料的分析,展现出了该算法的应用效果。
[Abstract]:In the field of seismic signal analysis and interpretation, wavelet transform has always been the focus and focus of people, so there are many wavelet transform time-frequency analysis methods, which are further improved on the basis of wavelet. From continuous wavelet transform, three parameter wavelet transform to S transform, and even generalized S transform, it is based on wavelet transform. It is also proved that these wavelet transform methods are more and more effective and the resolution is higher and higher when the signal is analyzed in time and frequency. But in order to meet the higher requirements, a new method of joint wavelet transform and recombination, synchronous extrusion wavelet transform, has emerged. Based on the principle of synchronous extrusion wavelet transform, this paper studies the properties of the algorithm, such as stability and reversibility, and finds that the reconstruction error is very small. At the same time, the influence of related parameters on the results of time-frequency analysis, such as wavelet mother function, parameter selection of wavelet mother function and wavelet threshold, is also studied. Through analysis and discussion, it is found that Morlet wavelet is generally chosen for seismic signal analysis, and the larger the center frequency is, the stronger the recognition ability of instantaneous frequency is. At the same time, the appropriate wavelet threshold can effectively suppress the interference of noise. Therefore, the reasonable setting of these parameters will greatly enhance the application effect of synchronous extrusion wavelet transform. Finally, aiming at the defects of synchronous extrusion wavelet transform in analyzing the signal whose instantaneous frequency change rate is not zero, the generalized synchronous extrusion wavelet transform is studied. However, the parameters in the algorithm are determined according to the characteristics of the signal to be analyzed. Therefore, its application is not strong. Continuous wavelet transform, three-parameter wavelet transform and synchronous extrusion wavelet transform are used for time-frequency analysis of analog signal and actual signal, respectively. The advantages of synchronous extrusion wavelet transform are compared. That is, with a higher time-frequency resolution. In the end, the amplitude response and frequency response of thin layer are studied, because they are characterized by time domain and frequency domain, respectively, and time-frequency analysis technology takes advantage of the two properties at the same time. Therefore, the synchronous extrusion wavelet transform is applied to the instantaneous spectrum analysis of thin interlayer, and the application effect of the algorithm is shown through the analysis of the actual seismic data.
【学位授予单位】：中国石油大学(华东)
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P631.4

【参考文献】