文献类型：期刊文章

作　　者：江沸菠[1,2] 戴前伟[2] 董莉[2,3]

机构地区：[1]物理与信息科学学院湖南师范大学,长沙410081 [2]地球科学与信息物理学院中南大学,长沙410083 [3]信息科学与工程学院湖南涉外经济学院,长沙410083

出　　处：《Applied Geophysics》

基　　金：supported by the National Natural Science Foundation of China(Grant No.41374118);the Research Fund for the Higher Education Doctoral Program of China(Grant No.20120162110015);the China Postdoctoral Science Foundation(Grant No.2015M580700);the Hunan Provincial Natural Science Foundation,the China(Grant No.2016JJ3086);the Hunan Provincial Science and Technology Program,China(Grant No.2015JC3067);the Scientific Research Fund of Hunan Provincial Education Department,China(Grant No.15B138)

年　　份：2016

卷　　号：13

期　　号：2

起止页码：267-278

语　　种：中文

收录情况：AJ、CSA、CSA-PROQEUST、CSCD、CSCD2015_2016、EBSCO、GEOREFPREVIEWDATABASE、IC、INSPEC、JST、PA、PROQUEST、SCI(收录号：WOS:000379578300005)、SCI-EXPANDED(收录号：WOS:000379578300005)、SCIE、SCOPUS、WOS、普通刊

摘　　要：Conventional artificial neural networks used to solve electrical resistivity imaging （ERI） inversion problem suffer from overfitting and local minima. To solve these problems, we propose to use a pruning Bayesian neural network （PBNN） nonlinear inversion method and a sample design method based on the K-medoids clustering algorithm. In the sample design method, the training samples of the neural network are designed according to the prior information provided by the K-medoids clustering results; thus, the training process of the neural network is well guided. The proposed PBNN, based on Bayesian regularization, is used to select the hidden layer structure by assessing the effect of each hidden neuron to the inversion results. Then, the hyperparameter αk, which is based on the generalized mean, is chosen to guide the pruning process according to the prior distribution of the training samples under the small-sample condition. The proposed algorithm is more efficient than other common adaptive regularization methods in geophysics. The inversion of synthetic data and field data suggests that the proposed method suppresses the noise in the neural network training stage and enhances the generalization. The inversion results with the proposed method are better than those of the BPNN, RBFNN, and RRBFNN inversion methods as well as the conventional least squares inversion.

关 键 词：Electrical resistivity imaging  Bayesian neural network  REGULARIZATION nonlinear inversion  K-medoids clustering  

分 类 号：P631.322]