【作者】 杨燕；

【导师】 刘建霞；

【作者基本信息】 太原理工大学 ， 电子科学与技术， 2015， 硕士

【摘要】 智能优化算法摆脱经典数学规划方法的束缚,从自然角度出发观构造直观的计算模型,解决优化问题。在数十年的研究与发展中,智能优化算法已成为各领域中解决问题的有力工具。近年来,智能算法开始涉足电磁领域,并被应用在天线设计和优化中,表现出卓越的性能,开始成为天线以及其他电磁部件设计中一种重要的方法。蚁群算法是一种模仿蚂蚁觅食机制的启发式智能算法,这种算法搜索速度快,适应性高,已经在通信、控制、管理等领域被广泛应用。本文主要研究嵌入遗传操作的蚁群算法及其在微带天线及光子晶体设计优化中的应用。天线作为无线通信系统中信息传输的重要部件,其性能的优良直接关系到整个通信系统通信质量的好坏。现代通信系统小型化、微型化的发展趋势,使得具有低剖面、小尺寸的微带天线成为各国学者研究的重点。针对微带天线的一些固有缺陷,研究人员不遗余力的对其进行优化和改善。光子晶体的出现及其在微波波段的成功应用,为微带天线的改进提供了一个新的思路,通过在微带天线上加载光子晶体,可以抑制或削弱微带天线基板上的表面波,大大提高微带天线的辐射效率。本文介绍了蚁群算法基本原理,在此基础上加入遗传操作,增加了解的多样性,提高了蚁群算法的全局搜索能力,将算法命名为遗传蚁群算法,并将这一算法应用在微带天线设计优化中。论文的主要内容如下：(1)遗传蚁群算法的研究。针对蚁群算法容易陷入局部最优的问题,将全局性较强的遗传算法融入蚁群算法中,通过测试与比较,证明了该算法在全局性、搜索效率等方面的优越性。(2)微带天线设计。将遗传蚁群算法同HFSS软件结合,用于U槽矩形微带天线和宽带微带天线的设计优化中,经过算法优化之后的微带天线性能有了显著的提高。(3)光子晶体微带天线设计。利用遗传蚁群算法和“像素法”设计共面紧凑型光子晶体单元,并将其应用在微带天线中,改进后的微带天线的性能有了明显的改善。(4)微带阵列天线去耦。利用遗传蚁群算法设计微带阵列天线的共面紧凑型光子晶体结构,通过加载光子晶体,使微带阵列天线单元间的互耦作用有了明显的削弱。更多还原

【Abstract】 Intelligent optimization algorithm gets rid of the bondage of classicalmathematical programming method and build a visualized calculation model tosolve the optimization problems according to the nature. In the decades ofresearch and development, it has become the one of the powerful tool to solvethe problems in many field. In the early1990s, intelligent optimizationalgorithm involved in electromagnetic filed. Since then, intelligent optimizationalgorithm begin to applied to design and optimization of antenna, and Shows theexcellent advantages in performance. Intelligent algorithm has become animportant method in design of antenna and other electromagnetic component.Ant colony optimization (ACO) algorithm which imitates the forgaing activitiesof ants is a kind of heuristic intelligent algorithm. The high search speed andefficiency make it widely used in communication engineering, automaticcontrol, management and other fileds. In this paper Genetic-Ant ColonyOptimization algorithm is studied and applied to design of microstrip antennaand photonic crystals. As an important part of wireless communication system, the performanceof antenna determine the performance of communication system. Thecharacteristics of Microstrip antenna which is low profile and small size cater tothe needs of the development trend of modern communication system which isminiaturization and micromation. Microstrip antenna become a global hotresearch field. For some inherent defects, the researchers spare no effortto improve the performance of microstrip antenna. The emergence of photoniccrystal and its successful application in microwave provides a new way to theimprovement of microstrip antenna. By loading photonic crystals on themicrostrip antenna, the surface wave in the substrate can be weakened, whichgreatly enhanced the radiation efficiency of antenna.In this paper Ant Colony Optimization (GACO) algorithm is proposed,named Genetic-Ant Colony Optimization algorithm. Through being embeddedin genetic operation, the diversity of Feasible solution of the algorithm isincreased, so the global search ability is enhanced. GACO is applied to designof microstrip antenna. The main content of the paper is as follows:(1)Genetic-Ant Colony Optimization algorithm. In view of theshortcoming of traditional Ant Colony algorithm which is easy to fall in to localoptimal, the improved algorithm GACO is proposed. Through six test functionsto prove the superiority of GACO. (2)Design of microstrip antenna. The GACO is combined with HFSSsoftware to used in design of U-slot rectangular micrstrip antenna andbraodband microstrip antenna. All the antennas get a better performance.(3)Design of crystal microstrip antenna. Use the GACO algorithm and“pixel” method to design a uncoplanar compact photonic crystal unit, andapplied the unit to microstrip antenna. The performance of the antenna isobviously improved.(4)Design of microstrip antenna array. Using GACO algorithm to designuncoplanar compact photonic crystal structure between the array unit. Byloading the photonic crystal, the mutual coupling effect is obviously weakened.更多还原