自适应时窗变化的相干体技术研究

发布时间：2018-04-18 09:31

本文选题：地震数据解释 + 相干体技术　；参考：《吉林大学》2015年硕士论文

【摘要】：寻找石油等地下矿藏一直是我国长期战略性任务。目前矿藏勘探的主要方式是通过对地震资料进行解释以了解地质构造。在地震资料解释中，断层识别和绘制是一个重要研究内容。相干体技术是地震断层识别的一个重要方法，其通过分析地震相邻道数据相似情况来以数据形式表示出地层结构的不连续性。目前，采用固定时窗的相干体技术的讨论较为常见。然而，地震数据的频带范围随时间变化而变化，若时窗较大，会采样到多个同相轴的地震信号，难以刻画细节；若时窗较小，无法识别完整的波形，较容易受干扰。为解决此问题，本文研究了一种时窗长度可随频带变化而变化的相干体技术。本文研究了地震数据解释中断层识别和绘制问题，提出了一种基于小波变换的自适应时窗相干体技术。首先，文中详细论述了相干体技术，尤其是第三代相干体技术。其次，研究了地震数据瞬时属性的提取方法，以此计算根据地震信号瞬时变化的时窗长度。文中主要利用希尔伯特变换方法与小波变换方法得到瞬时属性，并根据瞬时属性实时调整时窗长度来计算相干值，并以二维灰度图片形式进行比较、分析。针对地质数据体的三维特性，以OpenGL将二维结果置于三维空间，以显示相干体内部结构。最后以OpenMP并行策略对自适应时窗相干体技术进行实现和优化，以减少运行时间。本文利用小波理论，研究和改进了第三代相干体技术。文中通过计算地震数据瞬时属性来实时调整时窗长度，以此计算相干值，可以提高相干体计算的精确性。同时，文中采用并行策略对算法进行优化改进，提升了算法实现速度。实验结果表明，文中方法在断层识别精度和效率上取得了良好的效果，并在处理多噪声数据方面表现良好。本文工作为改进相干体技术提供了一种新的思路，以帮助地质工作人员对地震数据进行更准确地解释。
[Abstract]:It has been a long-term strategic task for our country to search for underground mineral deposits such as oil.At present, the main way of mineral exploration is to understand geological structure by interpreting seismic data.Fault identification and rendering is an important research content in seismic data interpretation.Coherent volume technique is an important method for seismic fault identification. By analyzing the similarity of seismic adjacent trace data, the discontinuity of stratigraphic structure can be expressed in the form of data.At present, the discussion of coherent volume technique using fixed time window is more common.However, the frequency range of seismic data varies with time. If the time window is large, it will sample many seismic signals in the same phase axis, so it is difficult to depict the details, and if the time window is small, it is difficult to recognize the complete waveform, so it is easy to be disturbed.In order to solve this problem, a coherent volume technique is studied, in which the length of time window varies with the frequency band.The problem of fault identification and rendering in seismic data interpretation is studied in this paper. An adaptive time-window coherent volume technique based on wavelet transform is proposed.Firstly, the coherent volume technique, especially the third generation coherent volume technique, is discussed in detail.Secondly, the method of extracting instantaneous attribute of seismic data is studied to calculate the time window length according to the instantaneous variation of seismic signal.In this paper, Hilbert transform method and wavelet transform method are used to obtain the instantaneous attribute, and the temporal window length is adjusted in real time according to the instantaneous attribute, and the coherent value is calculated and compared in the form of two-dimensional gray image.According to the three-dimensional characteristics of geological data volume, the two-dimensional results are placed in three-dimensional space by OpenGL to display the internal structure of coherent volume.Finally, the adaptive time-window coherent volume technique is implemented and optimized with OpenMP parallel strategy to reduce the running time.In this paper, the third generation coherent volume technique is studied and improved by wavelet theory.By calculating the instantaneous attribute of seismic data to adjust the length of time window in real time and calculate the coherence value, the accuracy of coherent volume calculation can be improved.At the same time, the parallel strategy is used to optimize and improve the algorithm.The experimental results show that the proposed method has achieved good results in fault recognition accuracy and efficiency, and has shown a good performance in dealing with multi-noise data.The work in this paper provides a new way of thinking for improving coherent volume technique to help geological workers to interpret seismic data more accurately.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P631.44

【参考文献】