同步挤压小波提取地震属性参数的研究

发布时间：2018-05-29 13:02

本文选题：地震属性 + 同步挤压小波变换　；参考：《西安理工大学》2017年硕士论文

【摘要】：地震属性不仅可以直接用来研究地球介质的岩性、构造等,也可以反映介质的品质因数等参数。但是在实际应用中,属性参数的测量精度以及抗干扰能力往往存在一定的局限性,针对这一问题,本文研究了同步挤压小波变换提取地震属性参数的方法。本文研究分为两部分:一是同步挤压小波变换对噪声的抑制,二是同步挤压小波变换对地震属性参数的提取。具体研究工作如下:(1)对窄带信号以及宽带信号分别运用小波变换、同步挤压小波变换进行了测试,研究了变换域的噪声分布规律;(2)对小波域的噪声抑制进行了探讨,研究了中值滤波、均值滤波和信息熵滤波在小波域中的去噪方法,并将三者滤波效果做了对比研究;(3)分别使用希尔伯特变换、同步挤压小波变换对地震信号进行了三瞬参数的提取研究,并通过对仿真信号和实际地震资料的处理验证了方法的有效性。实验结果表明,同步挤压小波变换是一种新的小波变换方法,可将时域信号转化成高分辨率的时频谱,再结合时频谱重排的思想,通过挤压任一中心频率附近区间值,从而得到同步挤压小波变换量值。其可将随机噪声压缩为点状噪声或者颗粒噪声,并聚集分布,基于噪声这一分布特点,从而选用合适的滤波方法,可达到很好的抑制噪声的目的;在此基础上,对地震信号进行瞬时属性的提取,相比于传统的希尔伯特变换,同步挤压小波可达到更高的精度以及更强的抗干扰能力。
[Abstract]:Seismic attributes can not only be directly used to study the lithology and structure of the earth's medium, but also reflect the parameters such as the quality factor of the medium. However, in practical application, the measurement accuracy and anti-interference ability of attribute parameters often have some limitations. In order to solve this problem, the method of extracting seismic attribute parameters by synchronous extrusion wavelet transform is studied in this paper. This paper is divided into two parts: one is the suppression of noise by synchronous extrusion wavelet transform, the other is the extraction of seismic attribute parameters by synchronous extrusion wavelet transform. The specific research work is as follows: (1) the narrowband signal and the wideband signal are tested using wavelet transform and synchronous extruding wavelet transform, respectively. The noise distribution law in transform domain is studied and the noise suppression in wavelet domain is discussed. The methods of median filtering, mean filtering and information entropy filtering in wavelet domain are studied. The extraction of three instantaneous parameters of seismic signal by synchronous extrusion wavelet transform is studied and the validity of the method is verified by processing the simulation signal and the actual seismic data. The experimental results show that synchronous extrusion wavelet transform is a new wavelet transform method, which can transform time domain signal into high resolution time spectrum, and combine with the idea of time spectrum rearrangement, by squeezing the interval value near any center frequency. The wavelet transform value of synchronous extrusion is obtained. The random noise can be compressed into point noise or particle noise and aggregated distribution. Based on the distribution characteristic of noise, a suitable filtering method can be used to suppress noise. Compared with the traditional Hilbert transform, the synchronous extrusion wavelet can achieve higher precision and stronger anti-jamming ability than the traditional Hilbert transform.
【学位授予单位】：西安理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P631.4

【参考文献】