基于神经网络的面波迭代反演应用研究
详细信息 本馆镜像全文    |  推荐本文 | | 获取馆网全文
摘要
根据滩浅海近地表结构特征,尝试利用地震记录中的面波进行近地表结构研究,以便了解滩浅海地区的近地表地层介质结构变化,为深层油气勘探提供准确的低降速带资料。针对面波频散反演已有方法存在的不足,引入一种基于BP神经网络的迭代反演方法对面波的频散曲线进行拟合迭代,用于反演预测滩浅海低降速带地层参数。由于神经网络具有很强的自学习、自适应、自组织和容错能力,它的反演预测能力非常强大,能够较精确地预测出所要求解的目标数据,同时结合传统迭代反演方法的优点,增强了该方法的反演预测能力。通过对滩浅海近地表结构模型试算,获得好的效果,同时进一步对实际记录进行了计算,也取得了比较满意的结果。
Because traditional methods applied in investigating surface-structure are restricted on paralic zone,according to the feature of surface-structure on paralic zone,the try is done to use the surface wave on seismic record to research the surface-structure and to offer deep exploration activity of weathering zone.In view of shortages in the methods applied in dispersion curve inversion of surface wave,an iterative inversion method based on BP(Back-propagation) artificial neural network is introduced to surface wave and it is used to predict the parameters of weathering zone on paralic zone.Combined with very strong self-learning,self-adapting,self-organizing and fault-tolerant ability of neural network,the prediction power of conventional iterative inversion method is enhanced effectively.By testing the model of paralic surface-structure,the good effect can be obtained.Moreover,applied to real data,the method still gives out satisfactory result.
引文
[1]徐淑合,刘怀山.滩浅海近地表地质结构研究[J].中国海洋大学学报,2003,33(增刊):37-42.
    [2]赵东,王光杰,王兴泰,等.用遗传算法进行瑞利波反演[J].物探与化探,1995,19(3):178-185.
    [3]Xia J,Miller R D,Park C B.Estimation of near-surface shear-wave velocity by inversion of Rayleigh waves[J].Geoghysics,1999,64(3):691-700.
    [4]欧阳联华,王家林,吴建生.利用面波频散反演场地参数的Occam法[J].物探化探计算技术,2003,25(1):1-4.
    [5]宋先海,肖伯勋,余才盛,等.瞬态瑞雷波反演横波的SVD算法及其应用[J].地质与勘探,2004,40(1):93-96.
    [6]崔建文.一种改进的全局优化算法及其在面波频散曲线反演中的应用[J].地球物理学报,2004,47(3):521-527.
    [7]阎平凡,张长水.人工神经网络与模拟进化计算[M].北京:清华大学出版社,2000.
    [8]冉启全,周雪峰,胡永乐.基于神经网络的多属性火成岩岩性反演技术[J].西南石油学院学报,2006,28(1):5-8.
    [9]Linden A,Kinderman J.Inversion of multi-layer nets[C]∥IJCNN.Proceedings of the International Joint Conference on Neural Networks.Washington DC:IJCNN,1989.
    [10]刘争平.人工神经网络在测井-地震资料联合反演中的应用研究[M].北京:科学出版社,2003.
    [11]闻新,周露,王丹力,等.MATLAB神经网络应用设计[M].北京:科学出版社,2000.
    [12]刘全稳.测井神经网络技术综述[J].石油地球物理勘探,1996,31(增刊1):64-69.
    [13]丛爽.面向Matlab工具箱的神经网络理论与应用[M].安徽合肥:中国科学技术大学出版社,1996.
    [14]杨成林.瑞雷波勘探[M].北京:地质出版社,1993.
    [15]何世聪.瑞雷波频散曲线提取方法的研究[D].湖北武汉:中国地质大学,2005.

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