| 直流电阻率测深二维自适应正则化反演 |
| |
| 引用本文： | 柳建新,彭艳华,刘海飞,孙丽影.直流电阻率测深二维自适应正则化反演[J].物探化探计算技术,2014(4):385-388. |
| |
| 作者姓名： | 柳建新 彭艳华 刘海飞 孙丽影 |
| |
| 作者单位： | 中南大学地球科学与信息物理学院;有色资源与地质灾害探查湖南省重点实验室;浙江省工程物探勘察院; |
| |
| 摘 要： | 直流电阻率测深二维反演中,正则化参数的选取影响反演结果分辨率及反演过程稳定性。利用主动约束平衡正则化因子,进行直流电阻率光滑约束最小二乘二维自适应反演,改善直流电阻率测深二维反演的分辨率与稳定性。在反演迭代过程中,正则化因子根据模型参数的空间展布函数进行自适应计算、正则化参数的自适应计算。模拟数据反演结果验证了该方法的有效性与可行性,反演结果能准确地反映地下模型的真实电性结构。
|
| 关 键 词： | 直流电阻率反演 自适应 主动约束平衡 正则化因子 模型分辨矩阵 展布函数 |
Two-dimensional self-adaptive regularization Inversion for DC resistivity data with active constraint balancing |
| |
| Institution: | LIU Jian-xin, PENG Yan-hua, LIU Hai-fei, SUN LI-ying (1. School of Geoscience and Info-Physics, Central South University, Changsha 410083,China 2. The institue of Engineering geophysical prospecting of Zhejiang Province,Hangzhou 310005,China 3. Key Laboratory of Non-ferrous Resources and Geological Detection,Ministry of Hunan Province,Changsha 410038,China) |
| |
| Abstract: | The determination of a suitable regularization parameter is necessary to achieve both resolution and stability in process of DC resistivity sounding inversion of two-dimension.In this paper,we present an inversion scheme for DC resistivity data u-sing self-adaptive regularization parameter with Active Constrained Balancing （ACB）approach.The regularization parameter varies spatially according to the spread function and is related to both the model parameters resolution matrix and the iteration number.The ACB approach assigns a higher value of regularization to low resolution.The varying regularization parameter ob-tained with the ACB approach is used to balance between the norm of data misfit and the norm of the model.The synthetic tests show that the inversion scheme with the ACB approach performs stable and convergence,and the inversion results can reflect accurately the subsurface geology structure. |
| |
| Keywords: | inversion of direct current resistivity self-adaptive active constraint balancing regularization parameter model parameter resolution matrix spread function |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
