首页> 外文学位 >Finite difference delay modeling for time domain integral equations of electromagnetics.
【24h】

Finite difference delay modeling for time domain integral equations of electromagnetics.

机译:电磁时域积分方程的有限差分时滞建模。

获取原文
获取原文并翻译 | 示例

摘要

Time domain integral equation (TDIE)-based methods for the electromagnetic scattering and radiation problems have many potential applications in the areas of high-resolution radar technology, electromagnetic pulse simulation studies, and target identification techniques. These applications could benefit from TDIE methods because of their combination of the strengths of integral equation methods and time domain methods. Specifically, as integral equation methods, they need only surface discretization for homogeneous scatterers, and as time domain methods, they can work for nonlinear problems and can analyze a band of frequencies in a single simulation.;Despite these advantages, TDIE methods have historically been inefficient and unstable, and therefore have not been applied broadly. This thesis develops an absolutely stable and accurate TDIE-based technique called the finite difference delay modeling (FDDM) method. In the FDDM method, the temporal discretization is realized by a mapping from the Laplace domain to the Z -domain based on a finite difference approximation derived from an ordinary differential equation solution method. Once the system is in the Z -domain, it can be inverse-transformed into a discrete time system and solved by marching-on-in-time. For Green's functions with simple Laplace domain expressions, the process can be carried out analytically. For other Green's functions or discretization schemes, a numerical method is employed to calculate the inverse Z -transform using trapezoidal rule and discrete Fourier transform (DFT).;The first FDDM method developed here computes scattering from perfect electric conductors (PECs). For the temporal discretization, first- and second-order finite difference approximations are used and are shown to be unconditionally stable. For open scatterers, there is a slowly growing, low frequency instability at later time steps because the electric field integral equation is blind to static solenoidal currents which generate no electric field. This problem can be solved by a loop-tree decomposition approach.;The second application of the FDDM scheme presented here computes the scattering from homogeneous dielectric bodies. Low frequency instability problems were avoided with another stabilization technique that augments the tangential field boundary condition equations with normal field boundary condition equations. In addition, the FDDM method was applied to dispersive scattering problems. Using FDDM, dispersive scattering is not much harder to model than non-dispersive scattering, though the kernels can be difficult to compute analytically. Thus, a numerical method is employed to compute the inverse Z -transform needed to discretize the kernel in time.;Finally, to get better temporal convergence, implicit Runge-Kutta based (IRK) based schemes are applied for the temporal discretization. The proposed technique maps a Laplace domain equation to a Z -domain equation using the Butcher tableau of the IRK scheme. A discrete time domain system is recovered by computing the inverse Z -transform numerically. The resulting technique is capable of third- or fifth-order accuracy in time, and is absolutely stable. Numerical results illustrate the accuracy and stability of the technique.
机译:基于时域积分方程(TDIE)的方法解决电磁散射和辐射问题,在高分辨率雷达技术,电磁脉冲仿真研究和目标识别技术领域具有许多潜在应用。这些应用可以从TDIE方法中受益,因为它们结合了积分方程方法和时域方法的优势。具体而言,作为积分方程方法,它们仅需要对均匀散射体进行表面离散化;作为时域方法,它们可以解决非线性问题并可以在单个仿真中分析频带。尽管有这些优势,但从历史上看,TDIE方法一直是效率低下且不稳定,因此尚未得到广泛应用。本文提出了一种绝对稳定和准确的基于TDIE的技术,称为有限差分时延建模(FDDM)方法。在FDDM方法中,时间离散是通过基于从常微分方程解法导出的有限差分近似值从拉普拉斯域到Z域的映射来实现的。一旦系统进入Z域,就可以将其逆变换为离散时间系统,并通过按时进行求解。对于具有简单Laplace域表达式的Green函数,可以通过分析来执行该过程。对于其他格林函数或离散化方案,采用数值方法使用梯形法则和离散傅里叶变换(DFT)来计算Z逆变换。此处开发的第一个FDDM方法是计算理想电导体(PEC)的散射。对于时间离散化,使用一阶和二阶有限差分近似,并且显示为无条件稳定。对于开放式散射体,在随后的时间步长处会缓慢增长,出现低频不稳定性,这是因为电场积分方程对不产生电场的静态螺线管电流视而不见。这个问题可以通过循环树分解法来解决。此处提出的FDDM方案的第二个应用是计算均质介电体的散射。使用另一种稳定技术避免了低频不稳定性问题,该稳定技术将切向场边界条件方程式扩展为法向场边界条件方程式。此外,FDDM方法应用于色散问题。使用FDDM,色散散射的建模并不比非色散散射困难,尽管内核可能难以进行分析计算。因此,采用一种数值方法来计算及时离散化内核所需的反Z变换。最后,为了获得更好的时间收敛性,将基于隐式Runge-Kutta(IRK)的方案用于时间离散化。所提出的技术使用IRK方案的Butcher表将Laplace域方程映射到Z域方程。通过数值计算逆Z变换可恢复离散时域系统。所产生的技术在时间上具有三阶或五阶精度,并且绝对稳定。数值结果说明了该技术的准确性和稳定性。

著录项

  • 作者

    Wang, Xiaobo.;

  • 作者单位

    University of Delaware.;

  • 授予单位 University of Delaware.;
  • 学科 Engineering Electronics and Electrical.
  • 学位 Ph.D.
  • 年度 2010
  • 页码 125 p.
  • 总页数 125
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号