【作者】 张倩；

【导师】 高原； 彭真明；

【作者基本信息】 电子科技大学 ， 电子与通信工程（专业学位）， 2016， 硕士

【摘要】 提高地震资料分辨率一直是地球物理勘探中一个十分重要的研究方向,薄储层预测研究更成为目前石油勘探学者们的研究热点。薄层反射系数反演经过十几年的发展,已经成为一种有效识别薄层的新方法。通过将地震资料变换到频域,选取优势频段进行反演,在频域求解目标函数可获得时间域的反射系数。反演得到的反射系数分辨率较高,可以有效的识别薄层。本文针对薄层反射系数反演方法开展了一系列研究,深入研究了薄层反射系数反演相关理论,建立了一套较完整的反演算法流程,主要工作和取得的成果包括:(1)研究了地震褶积模型,对地震褶积理论模型中地震子波,反射系数,地震记录分别进行了详细论述。重点论述了模型的时、频域表示以及地震记录在反演中做加窗傅里叶变换的原理。(2)讨论了不同分辨率极限准则下地震薄层的定义,研究了薄层的地震响应和增大反射系数偶分量权重对提高地震资料分辨率的作用。(3)研究了统计子波的提取方法,重点研究了两种子波提取的方法,常相位子波提取和最小相位子波提取。本文给出了算法的流程图并分别使用理论模型和实际数据进行测试,验证了算法的正确性。(4)推导了薄层反射系数反演的目标函数,从两个反射系数的目标函数开始推导,最终得到多个反射系数的目标函数,并给出薄层反射系数反演流程。(5)研究了薄层反射系数反演中的稀疏优化算法,论述了几种典型的反演优化算法,重点研究了基追踪去噪算法,给出了算法原理以及算法流程。(6)通过多种理论模型和实际数据的测试,验证了本文提出算法的正确性并获得较好的反演效果,同时测试了噪声对反演效果的影响。更多还原

【Abstract】 Improving the resolution of seismic data has been a very important research direction in the field of geophysical exploration, and the prediction of thin layers has become a hot research topic. Thin-bed reflectivity inversion is a recently developed effective method which could identify fine thin layers. Based on seismic data Fourier transform with windows to obtain frequency and domain information of seismic data inversion reflectivity. In the frequency domain, high resolution time domain reflectivity can be obtained, The effective identification of thin reservoir can be realized by this method.This paper for thin-layer reflectivity inversion method carried out a series of studies, in-depth study of the theory of thin bed reflectivity inversion, A relatively complete inversion algorithm processes is established. The main work and achievements include:(1) The seismic convolution model is studied, and the seismic wavelet, reflectivity and seismic record are discussed in detail. The time and frequency domain representation of the model and the principle of the Fourier transform in the inversion of seismic records are discussed in the paper.(2) The definition of seismic thin layer under the different resolution limit criterion is discussed, the effect of the seismic response and the influence of the odd and even parts of reflectivity on the resolution in inversion is studied.(3) Method of extracting statistical wavelet is studied, focuses on two methods wavelet extraction, the constant phase wavelet extraction and minimum phase wavelet extraction. This paper presents a flowchart of the algorithm and uses theoretical models and real data to verify the correctness of the algorithm.(4) The objective function of the inversion of thin layer reflectivity is derived. The objective function of two simple reflectivity and the objective function of multiple reflectivity are derived.Finally, the inversion process of thin layer reflectivity is presented(5) Sparse optimization algorithm in thin layer reflectivity inversion is studied, and several typical inversion optimization algorithms are discussed. Focus on the basis pursuit de-noising algorithm, the principle and algorithm flow are presented.(6) Through a variety of theoretical model and the real data test, verify the correctness of the proposed algorithm and the inversion effect, Finally the effect of noise on the inversion is tested.更多还原