An Improved Gravity Search Algorithm and Its Application in Planar Array Synthesis

改进的引力搜索算法及在面阵综合中的应用

¹ School of Communication and Information Engineering, Xi'an University of Science and Technology, Xi'an 710054, China
² School of Electronics and Information, Northwestern Polytechnical University, Xi'an 710072, China

Received: 20 September 2019

Abstract

An improved gravity search algorithm, adaptive gravity search algorithm (AGSA), is proposed to solve the problem that the gravity neutralization caused by the cumulative effect of particle inertia mass at the end of iteration, which will affect the optimization performance. An adaptive decay factor is designed, which can produce different gravitation values at different iteration stages of the algorithm and accelerate the mining ability of the algorithm at the later iteration stage. In order to enhance the memory ability of the algorithm, the influence of elite particles is added to the realization of the speed to expand the exploration ability. The improved algorithm is used to optimize uniform concentric ring array, the main lobe width optimized by the AGSA is 6.7°narrower and the side lobe level is 5.1 dB and 1.8 dB lower than the algorithm in the literature. It is clear that the pattern obtained by AGSA meets the desired pattern very well. Moreover, when the number of iterations is 2 000, the fitness value of the improved algorithm is increased by 30%. It can be seen that AGSA outperforms the algorithm in the literature in evolutionary speed and accuracy. Sparse concentric ring array also has the same optimization results. The effectiveness of the proposed improved algorithm in solving the array pattern synthesis is proved.

摘要

针对引力搜索算法在迭代过程中粒子惯性质量的累积效应造成的引力中和对优化性能的影响问题，提出了一种改进算法：自适应引力搜索算法。设计了一种随迭代次数自适应调整的衰减因子，提高了迭代后期算法的开采能力；在速度的计算过程中加入精英粒子，增强了粒子的记忆能力，算法的探索能力得以提高。将改进算法用于均匀同心圆环阵中，和文献中的算法相比，自适应引力搜索算法优化的主瓣宽度窄了6.7°、旁瓣电平分别低了5.1 dB和1.8 dB，更接近期望的方向图；平均收敛曲线的结果中，在迭代次数为2 000时，算法的适应度值提高了30%，收敛速度更快，优化精度更高；稀布同心圆环阵列也具有同样的优化效果，证明了所提改进算法在解决面阵方向图综合时的有效性。