-
研究一种可以高效求解半空间金属目标电磁散射积分方程方法, 电场积分方程适用于任意结构电磁问题分析, 但是生成的矩阵条件数大, 迭代求解收敛性差; 而磁场积分方程生成的矩阵条件数小, 迭代收敛性好, 但是仅能分析闭合结构问题, 本文采用了混合场积分方程方法, 同时具备电场积分方程的普适性与磁场积分方程的收敛性. 由于混合场积分方程中涉及格林函数的梯度项, 为了进一步加快计算效率, 本文引入了一种针对半空间格林函数的高效四维空间插值方法, 对组成半空间格林函数的索末菲积分进行列表和Lagrange插值, 以实现高效的迭代求解, 效率在传统混合场积分方程的基础上提高12.6倍. 数值结果表明, 该方法在保证精度的同时, 可以显著降低求解问题的时间.A new acceleration method is proposed for efficiently solving the problem of electromagnetic scattering from metal targets in half-space. The analysis of electromagnetic problems in any structure can be settled by the electric field integral equation. But the generated matrix condition number is large and the iterative solution has poor convergence. The number of the matrix condition generated by the magnetic field integral equation is small and iterative convergence is good. But only the closed structure problems can be worked out. The combined field integral equation is adopted because of the universality of the electric field integral equation and the convergence of the magnetic field integral equation. The gradient term of Green's function is involved in the integral equation of the mixed field. In order to further enhance the calculation efficiency, an efficient four-dimensional spatial interpolation method is introduced for half-space Green's function. Tabulation and lagrange interpolations are performed in the Sommerfeld integrals for the half-space Green's function. The improved efficiency can be 7.5 times higher than that of the traditional combined field integral equation(CFIE). Numerical results show that the computational time can be reduced significantly by the proposed method with encouraging accuracy.
-
Keywords:
- electromagnetic scattering/
- combined field integral equation/
- half-space Green's function/
- spatial interpolation/
- sommerfeld integral
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] -
模型 未知量 计算时间/s 计算内存/MB FEKO 传统CFIE 本文方法 FEKO 传统CFIE 本文方法 立方体 11682 1731 7560 1008 1229.0 1120.1 1251.9 slicy 7518 709 3204 432 533.6 545.8 557.0 船 14013 2541 18684 1476 1721.3 1613.6 1715.9 
下载:
导出CSV
-
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36]
下载:
计量
- 文章访问数:4165
- PDF下载量:80
- 被引次数:0