基于MP算法的快速地震信号谱分析
|
摘要
MP(Matching Pursuits)算法以其灵活的自适应性,取得了广泛的应用,但依然存在计算量大的应用瓶颈问题。为了解决传统MP算法运算量大的问题,首先对MP算法的原理进行了介绍,然后针对其计算量大的问题提出了一种基于遗传算法的快速算法,有效地降低了MP算法的运算量,较好地解决了运算量大的问题。最后利用MP算法对地震信号进行了谱分析,消除了Wigner算法中的交叉项,从而有效地揭示了地震信号的时频信息。结果表明,改进后的MP算法可以作为地震信号谱分析的有效算法,为烃类检测和储层描述等提供有地效的手段。
Because of the flexible self-adaptation,matching pursuits algorithm has been applied to many areas,but still has the problem of huge computational cost,so it is hard to be adopted in practice.To resolve the problem of huge computational cost,introduced the principle of matching pursuits algorithm firstly.A fast algorithm based on genetic algorithm was proposed secondly.This algorithm could decrease the amount of the computation effectively and solve the problem of huge computation.At the end,a segment of seismic signal was analyzed in time-frequency plane.In such a way the interference term in wigner distribution was eliminated,so the time-frequency information in the signal could be revealed.The result showed that the modified MP algorithm could be used as an efficient spectral analysis algorithm of seismic signal,and provided an effective means for hydrocarbon detection and reservoir description.
Because of the flexible self-adaptation,matching pursuits algorithm has been applied to many areas,but still has the problem of huge computational cost,so it is hard to be adopted in practice.To resolve the problem of huge computational cost,introduced the principle of matching pursuits algorithm firstly.A fast algorithm based on genetic algorithm was proposed secondly.This algorithm could decrease the amount of the computation effectively and solve the problem of huge computation.At the end,a segment of seismic signal was analyzed in time-frequency plane.In such a way the interference term in wigner distribution was eliminated,so the time-frequency information in the signal could be revealed.The result showed that the modified MP algorithm could be used as an efficient spectral analysis algorithm of seismic signal,and provided an effective means for hydrocarbon detection and reservoir description.
引文
[1]Mallat S,Zhang Z.Matching pursuit with ti me-frequencydictionaries[J].IEEE Trans.Signal Process,1993,41(12):3397-3421.
[2]周颖,王涛,冯焕清.匹配追踪算法及其在MI-EEG的应用[J].电路与系统学报,2005,10(4):37-41.
[3]张文耀.基于匹配跟踪的低位率语音编码研究[D].北京:中国科学院研究生院(软件研究所),2002.
[4]Castagna J P,Sun Shengjie.Comparisonof spectral decomposi-tion methods[J].Eage,2006,24:75-79.
[5]邵君.基于MP的信号稀疏分解算法研究[D].成都:西南交通大学,2006.
[6]陈发宇,尚永生,杨长春.Matching Pursuit方法综述[J].地球物理学进展,2007,22(5):1466-1473.
[7]雷英杰,张善文,李续武,等.MATLAB遗传算法工具箱及应用[M].西安:西安电子科技大学出版社,2005:11-13.
[8]张跃飞.基于稀疏分解的图像压缩[D].成都:西南交通大学,2006.
[9]陈发宇,杨长春.基于MP方法的地震信号快速分解算法[J].地球物理学进展,2007,22(6):1692-1697.
[10]马国田,梁昌洪.基于混合遗传算法的宽带阶梯阻抗变换器的优化设计[J].西安电子科技大学学报,1999,26(1):8-12.
[11]陈学华.时频分布与地震信号谱分析研究[D].成都:成都理工大学,2006.
[12]Castagna J P,Sun S,Siegfried R W.Instantaneous spectralanalysis:Detection of low-frequency shadows associated withhydrocarbons[J].The Leading Edge,2003,22:120-127.
[2]周颖,王涛,冯焕清.匹配追踪算法及其在MI-EEG的应用[J].电路与系统学报,2005,10(4):37-41.
[3]张文耀.基于匹配跟踪的低位率语音编码研究[D].北京:中国科学院研究生院(软件研究所),2002.
[4]Castagna J P,Sun Shengjie.Comparisonof spectral decomposi-tion methods[J].Eage,2006,24:75-79.
[5]邵君.基于MP的信号稀疏分解算法研究[D].成都:西南交通大学,2006.
[6]陈发宇,尚永生,杨长春.Matching Pursuit方法综述[J].地球物理学进展,2007,22(5):1466-1473.
[7]雷英杰,张善文,李续武,等.MATLAB遗传算法工具箱及应用[M].西安:西安电子科技大学出版社,2005:11-13.
[8]张跃飞.基于稀疏分解的图像压缩[D].成都:西南交通大学,2006.
[9]陈发宇,杨长春.基于MP方法的地震信号快速分解算法[J].地球物理学进展,2007,22(6):1692-1697.
[10]马国田,梁昌洪.基于混合遗传算法的宽带阶梯阻抗变换器的优化设计[J].西安电子科技大学学报,1999,26(1):8-12.
[11]陈学华.时频分布与地震信号谱分析研究[D].成都:成都理工大学,2006.
[12]Castagna J P,Sun S,Siegfried R W.Instantaneous spectralanalysis:Detection of low-frequency shadows associated withhydrocarbons[J].The Leading Edge,2003,22:120-127.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心