针对遗传算法的缺陷,文中首先提出了一种改进的结合局部搜索算法的混合遗传算法,并用5个测试函数对混合遗 传算法的性能进行了测试,充分说明了新算法的有效性。最后利用混合遗传算法对一种由6×6个单元构成的非周期大间 距阵列进行了综合,±15°扫描范围内最大副瓣电平小于-16 dB,扫描增益大于24 dB,达到了设计要求,并对最终结果 进行了分析,提出了改进的方法。

A modified genetic algorithm, named hybrid genetic algorithm with local search algorithm is proposed for overcoming the disadvantages of genetic algorithm. Five benchmark functions are used to validate the performance of the proposed algorithm. Hybrid genetic algorithm, is introduced in synthesis of an aperiodic array using large distance of elements and constituted by 6×6 units, maximum sidelobe level <-16 dB in the scanning range from -15°to 15°, scanning gain> 24 dB, to achieve the design requirements. The final result is analysed, and methods to improve result are proposed.