基于改进遗传算法的圆阵列方向图联合优化

为降低圆阵列方向图的峰值旁瓣电平，提出一种基于改进遗传算法的圆阵列方向图优化方法。将阵元位置和阵元权值作为联合优化变量，以最小化波束方向图峰值旁瓣为目标函数，采用遗传算法优化阵元位置和阵元权值，以增加变量的自由度，在采用双重选择机制的基础上，结合差分进化、内插/外推、单点交叉和多点交叉4种方式实现交叉变异。实验结果表明，该方法能降低陷入局部最优点的概率，具有较好的适应度和较快的收敛速度，使峰值旁瓣电平降低至?12.611 dB。

Joint Optimization of Circular Array Pattern Based on Modified Genetic Algorithm

In order to reduce the circular array pattern of the peak sidelobe level, this paper proposes a circular array pattern optimization method based on modified Genetic Algorithm(GA). This method makes the location of the array element and the coefficient as joint variables. By minimizing beam pattern peak sidelobe as the objective function, it not only can enhance the variables freedom degree but also can use GA to optimize the array element position and array element weights. In order to avoid premature convergence, differential evolution, interpolate, one-point crossover, multi-point crossover can be united based on the double choice mechanism. Experimental results show that this method can reduce the probability of getting into the local advantages, have good fitness and faster convergence speed, and make the peak sidelobe level decrease to ?12.611 dB.