电磁成像正演仿真与传感器参数优化设计  

Forward simulation for electromagnetic tomography and optimized design for sensor parameters

作　　者：王滨涛[1] 吴锡令[1] 王晓星[1] 

机构地区：[1]中国石油大学资源与信息学院测井研究中心,北京102249

出　　处：《地球物理学报》2009年第12期3142-3146,共5页Chinese Journal of Geophysics

基　　金：国家自然科学基金(60472019);中国石油天然气集团公司测井重点实验室项目资助.

摘　　要：应用有限元方法对电磁层析成像测量进行正演模拟,计算不同测量组合下各种典型流型的电势实部和虚部分布,考察电极个数、电极张角以及电极厚度对模拟测量响应的影响,并采用神经网络方法对传感器各结构参数进行优化设计.研究结果表明,介质分布对电势影响起主要作用,其中虚部分布特征更加明显;传感器各参数中电极数目对测量值影响较大,电极张角和电极厚度有一定影响.Firstly, FEM （Finite Element Method） is used to analyze forward problem and compute real potential and imaginary potential at different manifold distribution. Secondly, study the effect of the number of sensors, the angle and electrode thickness. Thirdly, the optimized parameters of EMT are gotten by neural network method. The parameters of sensor design are optimized by using neural network method. It can be found that the effect of potential is mainly affected by medium distribution, especially the imaginary potential distribution. The different number of sensors affects the measurement largely and the potential is affected by different angle and electrode thickness to some extent.

关 键 词：电磁层析成像 有限元方法 电势分布 优化设计 神经网络 

分 类 号：P631[天文地球—地质矿产勘探]