首页 | 本学科首页   官方微博 | 高级检索  
     

起伏地形三维激电连续介质模型快速反演
引用本文:刘海飞,柳建新,郭荣文,邓小康,阮百尧. 起伏地形三维激电连续介质模型快速反演[J]. 吉林大学学报(地球科学版), 2011, 41(4): 1212-1218
作者姓名:刘海飞  柳建新  郭荣文  邓小康  阮百尧
作者单位:1.中南大学信息物理工程学院,长沙410083;2.桂林理工大学广西地质工程中心重点实验室,广西 桂林541004;3.桂林理工大学地球科学学院,广西 桂林541004
基金项目:国家自然科学基金,广西科学基金,高等学校博士学科点专项科研基金,广西地质工程中心重点实验室开放基金,有色资源与地质灾害探查湖南省重点实验室项目
摘    要:为提高起伏地形三维激电数据的反演解释精度和处理效率,笔者开展了连续介质模型快速反演方法研究。针对起伏地形连续电性介质模型,提出了四面体单元电性参数分块线性连续变化的参数化方法。在反演中对模型参数施加光滑和背景约束信息,以提高反演的稳定性和分辨率。通过采用降低反问题的维数、压缩存储线性反演方程各矩阵的元素以及将互换原理和拟牛顿法相结合的方式计算偏导数矩阵等手段,可有效地加快反演的计算速度。最后,为验证反演方法的有效性,对2例地电模型进行反演试算,计算结果表明:反演耗费时间较少,仅迭代6次拟合差便趋于稳定,反演结果能较好地刻画异常体形态,编制的反演解释软件可用于实际生产。

关 键 词:勘查地球物理  起伏地形  三维激电数据  连续介质模型  快速反演  偏导数矩阵  
收稿时间:2010-10-22

Efficient Inversion of 3D IP Data for Continuous Model with Complex Geometry
LIU Hai-fei,LIU Jian-xin,GUO Rong-wen,DENG Xiao-kang,RUAN Bai-yao. Efficient Inversion of 3D IP Data for Continuous Model with Complex Geometry[J]. Journal of Jilin Unviersity:Earth Science Edition, 2011, 41(4): 1212-1218
Authors:LIU Hai-fei  LIU Jian-xin  GUO Rong-wen  DENG Xiao-kang  RUAN Bai-yao
Affiliation:1.School of Info-Physics and Geometrics Engineering, Central South University, Changsha410083, China;
2.Key Laboratory of Geological Engineering Centre of Guangxi Province, Guilin University of Technology, Guilin541004, Guangxi,China;
3.College of Earth Sciences, Guilin University of Technology, Guilin541004, Guangxi|China
Abstract:The study of efficient inversion method was carried out for continuous formation, targeting at improving the inversion precision and efficiency of 3-dimensional (3D) induced polarization (IP) data with waved geometry. Blocky continuously varied parameterization based on tetrahedral element was developed for the continuous changed resistivity model with fluctuated geometry. The stability and resolution of inversion was improved by including model smoothness and background prior information. The inversion efficiency was achieved via decreasing the dimension of inverse problem, memory compression of the matrix of linear equations, and calculation of sensitivity matrix with a combination of reciprocal theory and quasi-Newton method. Finally, the effectiveness of the inversion method was tested by two examples. The numerical results show that the inversion converges within 6 iterations and takes very little computational time. The inversion result can give a good interpretation of the anomalous targets. The inversion software developed can be used to interpret the real data.
Keywords:geophysical prospecting  complex geometry  3D induced polarization data  continuous model  efficient inversion  sensitivity matrix  
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《吉林大学学报(地球科学版)》浏览原始摘要信息
点击此处可从《吉林大学学报(地球科学版)》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号