【摘要】 首次采用免疫算法对稀疏阵列的方向图综合问题进行了研究。针对矩形平面稀布阵的结构特点,推导了克罗内克积形式的位置阵列因子这一新参数,减少了运算量,提高了运算速度;并对免疫算法进行了改进,通过采用自适应克隆和高斯变异等操作克服算法搜索的盲目性,提高了算法的搜索效率。针对阵元位置分布和激励幅度的特点依次应用二进制免疫算法和实数免疫算法来优化矩形平面稀疏阵列的旁瓣电平,表明了免疫算法的灵活性和有效性;采用免疫算法的优化结果要明显优于采用基本遗传算法的,表明了免疫算法具有更好的全局收敛特性。更多还原

【Abstract】 An immune algorithm is presented for the research on pattern synthesis in thinned array.Aiming at the structure of the rectangular plane thinned array,an array location factor is presented as a new factor with the form of Kronecker product.As a result,the amount of the computation is reduced while the speed of operation is improved.The immune algorithm is improved through the use of adaptive clone and Gauss immutation so as to overcome the blind search of the algorithm.As a result,searching efficiency is advanced.For optimizing the sidelobe of the rectangular plane thinned array,a binary immune algorithm and a real immune algorithm are applied to optimize element distribution and their excitation amplitude respectively.The results show the flexibility and effectiveness of the immune algorithm,and the optimal result applying immune algorithm is obviously better than the one using genetic algorithm.The proposed result shows a better characteristic of global convergence.更多还原