改进的萤火虫算法及在阵列天线中的应用

发布时间：2018-02-03 09:59

本文关键词： 群智能优化算法萤火虫算法维度变化阵列天线综合　出处：《杭州电子科技大学》2017年硕士论文　论文类型：学位论文

【摘要】：近年来,随着科学的进步和社会的快速发展,人们所面临的优化问题日益复杂化、大型化。因此寻求一个高效的、具有普遍适用性的智能优化算法具有十分重要的意义。群智能优化算法作为随机性优化算法的一个重要分支,因其较强的自适应性并可用来解决传统确定性优化算法难以解决的复杂优化问题,而受到众多学者的青睐,并广泛应用到各种工程领域,展现出其蓬勃的生命力。萤火虫算法作为一种提出来不久的群智能优化算法,虽然有诸如概念简单、易于实现、局部寻优能力好等优点,但仍然存在着在高维复杂问题上全局寻优能力不足、收敛速度缓慢等缺点。鉴于此,本文对萤火虫算法做出了改进,并将改进的萤火虫算法运用在天线的设计领域上。论文的主要研究成果如下:(1)从算法的基本原理以及数学模型两方面深入地研究了标准萤火虫算法,并通过和粒子群算法、人工蜂群算法两种常见的群智能优化算法进行仿真实验对比总结出萤火虫算法的优点与缺点,为算法的改进提供了思路。(2)为了提高萤火虫算法的全局寻优能力,以及加快收敛速度,本文对萤火虫算法做出的主要改进如下:首先,通过佳点集原理均匀初始化萤火种群,以提高初始解的质量使种群更加多样性。其次,对算法的可变参数步长因子α进行了研究分析,并提出了一种动态调整步长机制,以平衡算法的全局寻优和收敛速度。然后,通过引入全局最佳个体改变原始算法的位置更新公式,并对最佳粒子的位置基于维度的变化更新,以此使算法更加容易跳出局部最优,增强了全局寻优的能力。最后,通过7个标准测试函数的测试表明,本文改进的算法不仅收敛速度快而且寻优精度高。(3)将改进后的萤火虫算法应用到了阵列天线的方向图综合问题上,仿真测试结果表明,改进后的算法相对于传统的优化方法以及标准萤火虫算法都有较好的优化效果。
[Abstract]:In recent years, with the progress of science and the rapid development of society, the optimization problem that people are facing is becoming more and more complicated and large-scale. Intelligent optimization algorithm with universal applicability is of great significance. Swarm intelligence optimization algorithm is an important branch of stochastic optimization algorithm. Because of its strong adaptability and can be used to solve complex optimization problems which are difficult to solve by traditional deterministic optimization algorithm, it is favored by many scholars and widely used in various engineering fields. Firefly algorithm as a proposed swarm intelligence optimization algorithm, although it has such advantages as simple concept, easy to implement, good local optimization ability and so on. However, there are still shortcomings such as insufficient global optimization ability and slow convergence rate in complex high-dimensional problems. In view of this, this paper improves the firefly algorithm. And the improved firefly algorithm is applied in the field of antenna design. The main research results in this paper are as follows: 1) the standard firefly algorithm is deeply studied from two aspects: the basic principle of the algorithm and the mathematical model. And through comparing with particle swarm optimization and artificial bee swarm algorithm, two common swarm intelligence optimization algorithms are compared to sum up the advantages and disadvantages of the firefly algorithm. In order to improve the global optimization ability of the firefly algorithm and speed up the convergence of the algorithm, the main improvements made in this paper are as follows: first. In order to improve the quality of initial solution and make the population more diverse, the optimal point set principle is used to initialize the fluorescence population uniformly. Secondly, the variable parameter step factor 伪 of the algorithm is studied and analyzed. A dynamic step size adjustment mechanism is proposed to balance the global optimization and convergence speed of the algorithm. Then, the position updating formula of the original algorithm is changed by introducing the global best individual. The location of the best particle is updated based on the change of dimension, which makes it easier for the algorithm to jump out of the local optimum and enhance the ability of global optimization. Finally, the test results of seven standard test functions show that. The improved algorithm not only converges fast but also has high optimization accuracy. The improved firefly algorithm is applied to the pattern synthesis of array antenna. The simulation results show that the improved algorithm can be applied to the pattern synthesis of antenna array. The improved algorithm is better than the traditional optimization method and the standard firefly algorithm.
【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;TN820.15

【参考文献】