基于EMD和小波变换的核磁测井回波信号去噪研究

发布时间：2018-01-11 09:12

  本文关键词：基于EMD和小波变换的核磁测井回波信号去噪研究　出处：《东北石油大学》2014年硕士论文　论文类型：学位论文

  更多相关文章： 核磁共振测井 小波变换 经验模态分解 端点效应 去噪

【摘要】：核磁共振测井技术能够测得丰富的岩层信息，从而对地层进行准确的评价。但是核磁共振测井产生的回波信号是非常微弱的，伴随着大量的噪声干扰，导致回波信号的信噪比较低。因此，研究回波信号的去噪，提取更多的有用信息是核磁共振测井中非常重要的步骤。本文的主要工作如下： 首先，详细介绍了小波去噪的理论，应用MATLAB工具对仿真信号进行小波阈值去噪，分别研究了小波阈值去噪中阈值函数的选取，小波基函数的选取和小波软阈值、硬阈值的选取对去噪效果的影响。分析了信号和噪声的模极大值在尺度间的传播特性，用信号仿真对最大分解层次进行了分析，讨论了小波阈值去噪和小波模极大值去噪的优缺点。并运用小波去噪方法对现场采集的核磁共振回波信号进行去噪研究。 其次，研究了经验模态分解基本理论和基于经验模态分解算法的去噪过程。针对经验模态分解过程产生端点效应问题，提出了本文基于局部极值延拓和端点判断的方法。通过仿真实验证明了该方法能够有效的抑制端点效应，并将基于该方法的经验模态分解应用在回波信号的去噪中，得到了较好的效果。 最后，研究了核磁共振测井中回波信号噪声产生的因素。根据小波去噪理论中阈值和小波基函数选取问题，经验模态分解去噪粗糙，丢失有用信息等问题，结合两者的优点，提出了基于经验模态分解和小波变换的联合去噪方法，通过对现场采集的原始回波信号，应用小波阈值法，经验模态分解算法和联合去噪方法的结果进行对比分析，，得到融合的算法获取的信息更多，去噪效果更好，进一步提高了核磁共振测井中回波信号的有用信息含量。
[Abstract]:Nuclear Magnetic Resonance logging (NMR) technology can obtain abundant information of strata and evaluate the formation accurately, but the echo signal generated by NMR logging is very weak, accompanied by a large number of noise interference. Therefore, it is very important to study the denoising of echo signal and extract more useful information. The main work of this paper is as follows: Firstly, the theory of wavelet de-noising is introduced in detail, and the wavelet threshold denoising of simulation signal is carried out by using MATLAB tool, and the selection of threshold function in wavelet threshold de-noising is studied respectively. The influence of wavelet basis function selection and wavelet soft threshold and hard threshold selection on the denoising effect is analyzed. The propagation characteristics of the modulus maximum of signal and noise between scales are analyzed. The maximum decomposition level is analyzed by signal simulation. The advantages and disadvantages of wavelet threshold denoising and wavelet modulus maximum de-noising are discussed. Secondly, the basic theory of empirical mode decomposition and the denoising process based on empirical mode decomposition algorithm are studied. A method based on local extremum continuation and endpoint judgment is proposed in this paper. The simulation results show that the method can effectively suppress the endpoint effect. The empirical mode decomposition based on this method is applied to the denoising of echo signal, and good results are obtained. Finally, the factors of echo signal noise in NMR logging are studied. According to wavelet denoising theory, the selection of threshold and wavelet basis function, the rough denoising by empirical mode decomposition, the loss of useful information and so on. Combining the advantages of the two methods, a combined denoising method based on empirical mode decomposition and wavelet transform is proposed. The wavelet threshold method is applied to the original echo signal collected in the field. The results of empirical mode decomposition algorithm and joint denoising method are compared and analyzed. The fusion algorithm can get more information and better denoising effect. The useful information content of echo signal in NMR logging is further improved.
【学位授予单位】：东北石油大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：P631.81;TN911.4

【参考文献】

中国期刊全文数据库 前7条

1 池越;周亚同;;基于小波变换的数字水印算法研究与实现[J];电子制作;2013年05期

2 秦亚辉;冯景辉;陈立定;;基于小波变换的信号去噪方法研究[J];信息技术;2010年01期

3 管玲玲;柳忠校;张善心;;基于小波的信号分析[J];计算机与现代化;2011年03期

4 傅婷;陈莹;;核磁共振测井技术现状[J];矿物学报;2013年S2期

5 唐玲;程建;;小波变换在语音信号处理上的应用[J];四川理工学院学报(自然科学版);2012年01期

6 王文圣,丁晶,向红莲;小波分析在水文学中的应用研究及展望[J];水科学进展;2002年04期

7 张照煌;丁显;刘曼;曾菊瑛;;基于小波变换的风电机组传动系统故障诊断与分析[J];应用基础与工程科学学报;2011年S1期

中国博士学位论文全文数据库 前1条

1 王婷;EMD算法研究及其在信号去噪中的应用[D];哈尔滨工程大学;2010年

本文编号：1408926