基于动态Kriging的CBPSO算法在天线设计中的应用
Cultural-Particle Swarm Optimization Algorithm Based on Dynamic Kriging Model for Antenna Design
-
摘要: 针对天线优化设计通常涉及高度非线性的问题,传统优化算法往往无法获得全局最优解,在此研究背景下,引入文化粒子群优化算法(Cultural Based PSO Algorithm,CBPSO);针对高频电磁仿真软件(HFSS)仿真计算量大、耗时长的问题,引入Kriging 模型替代费时的仿真计算,并通过动态更新的方法提高模型精度,提出了基于动态Kriging 模型的文化粒子群算法与HFSS 联合仿真优化设计方案。将该方案应用于WLAN/ WiMAX 多频带天线优化设计,测试结果表明,所设计的天线在2.4~3.0GHz、3.3~3.8GHz、5.1~6.0GHz 频段内回波损耗小于-10dB,覆盖了WLAN/ WiMAX 所有频段,为复杂天线结构的优化设计提供了一定的参考。Abstract: The traditional optimization algorithms might fail to obtain the global optimal solution for highly nonlinear optimization problem in antenna design, so a cultural based particle swarm optimization (CBPSO) algorithm is induced, and by constructing Kriging model to replace time-consuming HFSS simulation evaluation, the model is updated to improve the approximation accuracy. Proposed a optimizing project based on the combination of dynamic Kriging model-CBPSO with high frequency struc-ture simulator (HFSS), use this project to design a WLAN/ WiMAX antenna, which covers the frequency band from 2.4 to 3.0GHz,3.3 to 3.8GHz,5.1 to 6.0GHz with -10dB return loss. It's proved that this project can be used to optimize the complicated structure antenna.