改进遗传算法在双弯曲反射面天线设计中的应用

摘要

将改进的遗传算法应用于双弯曲反射面天线的设计,旨在实现其低副瓣、宽带、低交叉极化及俯仰面波瓣赋形的指标要求.根据该类天线的设计经验,简化了优化目标,构建了一种新的适应度函数.在优化过程中,基于实数编码,初始种群尽量分散,采用精英选择与轮盘赌选择相结合的方法和改进的变异方法生成新一代种群,改善了收敛速度,克服了早熟现象.应用实例表明,基于改进遗传算法的优化设计结果达到了预期要求,与基于传统方法的计算结果相比,在波瓣赋形、交叉极化、带内指标一致性上有了较大的改善.%The improved genetic algorithm is used to design the doubly curved reflector antenna to satisfy the requirements of low sidelobe,wide band,low cross polarization isolation and the shaped pattern in elevation plane.By the design experience of this kind of antenna,the optimization targets is simplified and a new fitness function is constructed,as well as the whole excellent result is achieved.The optimization process is based on real code,in which the individuals in the original population differ widely.To create new population for next generation,the combination of elite selection and roulette selection is adopted,and the improved aberrance is also introduced,which improves the convergency performance and overcomes the precocious phenomenon.Finally,the application of an example is presented.The results show that the patterns of the far field satisfy the anticipate targets of the design.The shaped pattern and all the parameters based on the improved genetic algorithm are compared with those based on traditional method.Simulation results indicate that the method is effective and superior.