混合粒子群算法在阵列天线综合中的应用

发布时间：2018-06-12 18:37

本文选题：粒子群优化算法 + 细菌群体趋药性算法　；参考：《杭州电子科技大学》2017年硕士论文

【摘要】：由于实际优化问题情况复杂,传统优化方法对优化问题的依赖性强,在解决复杂、困难的优化问题时,往往具有较大的局限性;因此优化效果好、可用性强的群体智能算法获得发展,并被广泛用于自动化控制、模式识别、人工智能等各个领域。本文主要研究了群体智能算法中的粒子群优化算法(Particle Swarm Optimization,PSO),将其与细菌群体趋药性算法(Bacterial Colony Chemotaxis optimization,BCC)相结合,提出了一种混合粒子群算法—Particle Swarm Optimization and Bacterial Colony Chemotaxis optimization(PSOBCC),并将其应用于阵列天线进行降低旁瓣电平和生成深零点。论文的主要研究成果如下:(1)对粒子群算法的基本概念、实现方式、缺陷以及改进方式进行分析描述,进一步阐述了该算法的研究现状和发展趋势。(2)为了提高算法搜索速度,本文改变了粒子群算法的更新公式,只保留位置项进行迭代更新,并重新设置了惯性权重和学习因子的取值;同时为了提高算法的收敛精度,引入细菌群体趋药性算法进行局部搜索。整个优化过程中,对全局最优值进行随机扰动,并提出了精英替换策略。(3)对优化算法的一些常用测试函数进行研究,并将算法用单峰测试函数、多峰测试函数、经过旋转平移的经典测试函数这三类测试函数分别进行测试,并与一些最新的和经典的算法进行对比。(4)将本文的混合粒子群算法应用于阵列天线的方向图综合中,针对阵列天线中的低旁瓣和深零点进行优化,并取得了较好的结果。
[Abstract]:Because the actual optimization problem is complex and the traditional optimization method is strongly dependent on the optimization problem, it often has great limitations in solving the complex and difficult optimization problem, so the optimization effect is good. Swarm intelligence algorithms with high availability have been developed and widely used in automation control, pattern recognition, artificial intelligence and other fields. In this paper, particle swarm optimization (PSO) algorithm in swarm intelligence algorithm is studied, which is combined with bacterial colony chemotaxis algorithm (Bacterial Colony Chemotaxis optimization BCCs). A hybrid particle swarm optimization algorithm (-Particle Swarm Optimization and Bacterial Colony Chemotaxis optimization PSOBCC) is proposed and applied to array antennas to reduce sidelobe level and generate deep zeros. The main research results of this paper are as follows: (1) analyzing and describing the basic concept, implementation, defect and improvement of PSO, and further expounding the research status and developing trend of PSO.) in order to improve the search speed of PSO, In this paper, the updating formula of particle swarm optimization algorithm is changed, only the position term is reserved for iterative updating, and the values of inertia weight and learning factor are reset, meanwhile, in order to improve the convergence accuracy of the algorithm, This paper introduces the bacterial population drug-seeking algorithm to carry on the local search. In the whole optimization process, the global optimal value is randomly perturbed, and the elite substitution strategy is proposed. Some common test functions of the optimization algorithm are studied, and the single-peak test function and the multi-peak test function are used in the algorithm. The three kinds of test functions are tested respectively after rotation and translation, and compared with some new and classical algorithms, the hybrid particle swarm optimization algorithm is applied to the pattern synthesis of array antenna. The low sidelobe and deep zero of array antenna are optimized, and good results are obtained.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP18;TN820

【参考文献】