基于粒子群算法的高斯过程建模在天线优化上的应用研究

发布时间：2019-03-18 21:23

【摘要】：当对天线进行优化设计时,可以结合电磁仿真软件HFSS和粒子群优化算法予以实现,但是调用HFSS评估粒子群算法的适应度时需要花费大量的时间,同时也对计算机性能有较高的要求,从而对复杂结构的天线设计造成困难。本文针对多种用途和频段的天线,将高斯过程建模方法融合到粒子群优化算法中,对天线进行了优化设计,达到设计指标要求;同时和以往粒子群算法与HFSS软件相结合的方法进行了比较,证明高斯过程-粒子群联合算法在优化时间上具有极大的优势。本文主要的研究工作如下:1、介绍了高斯过程模型的建立方法及评估方法,以矩形侧馈微带天线模型为例,表明高斯过程模型预测该天线的频率特性具有一定的精确性。2、介绍了高斯过程-粒子群联合算法的思路,从时间的角度将该方法与调用电磁仿真软件HFSS作为粒子群算法适应度评价方案的方法进行了对比,说明了高斯过程-粒子群联合算法的优势。3、对印刷偶极子天线的频率特性进行高斯过程建模,解决了有带宽要求的天线频率特性建模问题,并将此高斯过程模型融合到粒子群算法中去,对印刷偶极子天线某些结构的尺寸进行优化设计,大幅减少了优化设计所需时间。4、对GPS北斗双模微带天线的频率特性进行高斯过程建模,解决了较窄的频带范围内对特殊频率点有要求的天线频率特性建模问题,建立起精确度比较高的高斯过程模型,融合到粒子群算法中,对该天线结构尺寸进行了优化设计,大幅减少了天线优化设计所需时间。5、对双脊喇叭天线的频率特性进行高斯过程建模,解决了宽频带天线频率特性建模问题,建立起高精度高斯过程模型,融合到粒子群算法中,对双脊喇叭天线结构尺寸进行了优化设计,大幅减少了该天线优化设计所需时间。
[Abstract]:When the antenna is optimized, it can be realized by combining electromagnetic simulation software HFSS and particle swarm optimization algorithm, but it takes a lot of time to use HFSS to evaluate the fitness of particle swarm optimization algorithm. At the same time, there is a high demand for computer performance, which makes the antenna design of complex structure difficult. In this paper, Gao Si process modeling method is integrated into particle swarm optimization (PSO) algorithm for antenna with various uses and frequency bands, and the antenna is optimized to meet the requirements of the design index. At the same time, the method of combining particle swarm optimization algorithm with HFSS software is compared, and it is proved that the Gao Si process-particle swarm optimization algorithm has great advantages in the optimization time. The main research work in this paper is as follows: 1. The establishment and evaluation methods of Gao Si process model are introduced. Taking the rectangular side-fed microstrip antenna model as an example, it is shown that Gao Si process model has certain accuracy in predicting the frequency characteristics of the antenna. This paper introduces the idea of Gao Si process-Particle Swarm Optimization Joint algorithm, and compares this method with the method of using electromagnetic simulation software HFSS as the fitness evaluation method of Particle Swarm Optimization algorithm from the point of view of time. The advantages of the Gao Si process-particle swarm joint algorithm are illustrated. 3, the Gao Si process modeling for the frequency characteristics of printed dipole antennas is carried out, which solves the problem of antenna frequency characteristics modeling with bandwidth requirements. The Gao Si process model is incorporated into the particle swarm optimization algorithm to optimize the size of some structures of printed dipole antenna, which greatly reduces the time required for optimization design. 4, The frequency characteristics of GPS dual-mode microstrip antenna are modeled by Gao Si process, which solves the problem of antenna frequency characteristic modeling for special frequency points in a narrow band range, and sets up a Gao Si process model with high accuracy. The structure size of the antenna is optimized by the particle swarm optimization algorithm, which greatly reduces the time required for the optimization design of the antenna. 5. The frequency characteristics of the antenna with double ridged horn are modeled by Gao Si process. The problem of frequency characteristic modeling of broadband antenna is solved. A high precision Gao Si process model is established and incorporated into particle swarm optimization algorithm. The structural size of the antenna with double ridged horn is optimized, which greatly reduces the time required for the optimum design of the antenna.
【学位授予单位】：江苏科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN820;TP18

【参考文献】