基于序列二次规划方法的高密度电阻率反演  

High density electrical resistivity inversion based on sequential quadratic programming method

作　　者：伍开江[1] 周启友[1] 

机构地区：[1]南京大学地球科学系,江苏南京210093

出　　处：《地球科学与环境学报》2005年第3期63-67,共5页Journal of Earth Sciences and Environment

基　　金：国家自然科学基金项目(40371021)

摘　　要：针对高密度电阻率成像法反演计算中存在的困难,探讨了基于序列二次规划(SQP)方法建立可行的电阻率反演算法的问题。在对三维点电源二维地电体电位场模型有限元法正演计算的基础上,建立了基于SQP方法的高密度电阻率成像法反演算法。根据勘测中测量电极间距的不同,提出按不同的电极间距分别建立优化模型进行参数优化。通过对模拟电阻率模型和实际观测资料的反演计算,表明该算法进行高密度电阻率反演是可行的,其具有对初始模型无特殊要求、收敛速度快的特点。这一反演算法可望在高密度电阻率成像的反演计算中得到应用。This paper developed a feasible inverse algorithm based on sequential quadratic programming （SQP） method to deal with the inversion problems of high density electrical resistivity tomography prospecting. Based on the forward calculation of electrical potential for three-dimensional point source and two-dimensional earth with finite element method, an inversion algorithm was built with SQP method, According to the difference of electrode intervals, the objective function of parameter optimization was formed respectively. Applications of the algorithm to artificial resistivity models and real data indicated that it is feasible in high density electrical resistivity inversion and has the advantages of no particular constraints to the initial parameter and high convergence speed. So this algorithm has great potential applications in the high density electrical resistivity tomography prospecting.

关 键 词：电阻率 反演 有限元法 参数优化 SQP方法 

分 类 号：P631.322[天文地球—地质矿产勘探]