基于实数编码遗传算法的赋形反射面天线设计  

Optimal Design of Shaped Reflector Antennas Based on Real Coded Genetic Algorithm

作　　者：张新刚[1] 吴刚[1] 钟鹰[1] 

机构地区：[1]中国空间技术研究院西安分院,西安710100

出　　处：《微波学报》2011年第5期23-27,54,共6页Journal of Microwaves

摘　　要：为了使星载天线的远场方向图与服务区的形状相匹配,并在服务区内获得高的增益和好的交叉极化隔离度,采用改进的实数编码遗传算法对其进行了优化设计。在优化过程中,以主极化增益和交叉极化隔离度作为优化目标,参考目标规划法来构造适应度函数;在群体选择的基础上采用差分进化方式、内插/外推和基因交换3种方法相结合来产生新个体,并采用自适应变异概率,来保证新一代染色体模式的多样性。最后以一个工作在Ku频段的赋形反射面天线为例,对其形状进行了优化设计。数值仿真结果表明,不仅天线的方向图与服务区形状匹配度很高,而且主极化增益和交叉极化隔离度都满足特定的设计要求,充分验证了所提算法的可行性和有效性。To make the far field pattern of the antenna mounted on satellite to match the coverage area,and obtain the high gain and the finer cross polarization isolation,the application of the improved real coded genetic algorithm in the design of a single offset shaped reflector antenna is addressed.In the optimization process,the gain and the cross polarization isolation are regarded as the optimization targets and the fitness function is constructed based on the goal programming.To guarantee the diversity of the chromosome,a hybrid method composed of difference evolution manner,insert/extrapolate and gene exchange is used to generate the new individuals and the adaptive aberrance probability is adopted.Finally,the numerical simulation results of a shaped reflector antenna at Ku-band are presented.The results show demonstrate that the pattern of the far field matches the shape of the coverage area very well and all the parameters are consistent with the targets of design,which fully validate the feasibility and validity of the proposed procedure.

关 键 词：遗传算法 赋形波束 反射面天线 优化 

分 类 号：TN823.27[电子电信—信息与通信工程]