摘要
提出了一种将标准遗传算法和确定性方法相结合的混合遗传算法，并应用该方法对相对带宽为１００％的宽带阶梯阻抗变换器进行优化设计，克服了标准遗传算法效率太低及确定性方法易收敛于局部极小点的缺点．分别对负载阻抗为纯实数和复数的两种情况进行优化设计表明：当负载为纯电阻时，混合遗传算法的计算结果与Ｃｈｅｂｙｓｈｅｖ综合所得结果基本一致；当负载为复阻抗时，混合遗传算法所得结果优于传统的综合方法．
A hybrid genetic algorithm (HGA) composed of the standard genetic algorithm (SGA) and the decisive optimal method is proposed, and a broadband stepped impedance transformer with 100% relative band width is designed optimally by employing the presented algorithm. HGA has a higher search effectiveness compared with SGA, and can lead to global convergence, unlike decisive optimal methods which may lead to local convergence. The transformer is designed respectively in the case of a real impedance load and a complex one. The design results are presented, which show that HGA agrees with Chebyshev synthesis method if the load is a resistor, and is better than the traditional network synthesis method if the load is a complex impedance.
出处
《西安电子科技大学学报》
EI
CAS
CSCD
北大核心
1999年第1期8-12,共5页
Journal of Xidian University
关键词
混合遗传算法
阻抗变换器
最优化
设计
hybrid genetic algorithm impedance transformer optimization
