基于FEM-MEI和正余弦算法的二维电磁成像方法  

A two-dimensional electromagnetic imaging method based on FEM-MEI and sine-cosine algorithms

作　　者：贾瀚钦 高红伟 郝凯子 贾祖朋[3] JIA Hanqin;GAO Hongwei;HAO Kaizi;JIA Zupeng(School of Integrated Circuits and Electronics,Beijing Institute of Technology,Beijing 100081,China;Beijing Simulation Center,Beijing 100854,China;Institute of Applied Physics and Computational Mathematics,Beijing 100094,China)

机构地区：[1]北京理工大学集成电路与电子学院,北京100081 [2]北京仿真中心,北京100854 [3]北京应用物理与计算数学研究所,北京100094

出　　处：《南京邮电大学学报（自然科学版）》2024年第4期68-76,共9页Journal of Nanjing University of Posts and Telecommunications：Natural Science Edition

基　　金：国家自然科学基金(62171027);北京理工大学青年教师学术启动计划资助项目。

摘　　要：针对二维介质目标的电磁成像问题,将正余弦算法(Sine Cosine Algorithm,SCA)与有限元方法(Finite Element Method,FEM)和不变性测试方程(Measured Equation of Invariance,MEI)进行结合提出一种新的成像方法。将FEM与MEI进行结合求解二维介质目标的电磁散射正问题,即求解Helmholtz方程。其中,MEI保证边界截断的精度,FEM适用于复杂介质目标的准确模拟。对于电磁散射逆问题,引入SCA并加以改进提出一种新的重构方法。该方法采用等效原理与格林函数的渐近式求得远区散射场,以测量的散射场和计算的散射场最大偏差为目标函数,采用改进的SCA优化介质参数,使目标函数达到最小值,以此重构散射体。为提高计算效率,采用MPI算法进行并行计算。文中采用基准函数展示了改进的SCA算法的快速收敛性,并采用非规则的均匀介质柱目标验证了成像方法的正确性。For electromagnetic imaging of two-dimensional dielectric targets,this paper proposes a new imaging method via combining the sine-cosine algorithm(SCA)with the finite element method(FEM)and the measured equation of invariance(MEI).FEM and MEI are combined to solve the forward electromagnetic scattering problem of two-dimensional dielectric targets,namely the Helmholtz equation.MEI guarantees the accuracy of the boundary truncation,and FEM is suitable for accurate modeling complex dielectric targets.For the electromagnetic inverse scattering problem,a new reconstruction method is proposed based on an improved version of the sine-cosine algorithm.The scattering fields in the far region are obtained using the equivalence principle and the asymptotic expression of the Greens function.The maximum deviation between the measured scattering fields and the calculated scattering fields is taken as the objective function.The medium parameters are optimized by sine-cosine algorithm to minimize the objective function,and the scatterer is reconstructed.Furthermore,the MPI parallel computing is adopted to improve computing efficiency.Finally,a reference function is used to demonstrate the fast convergence of the improved sine-cosine algorithm,and the correctness of the imaging method is verified by an irregular homogenous dielectric column object.

关 键 词：正余弦算法 有限元方法 MEI方法 电磁散射 二维成像 

分 类 号：TN011[电子电信—物理电子学] O451[理学—无线电物理]