基于改进粒子群算法的标签天线结构参数多目标优化设计

Multi-objective Optimization Design of Tag Antenna Structure ParametersBased on Improved PSO Algorithm

中文关键词:  射频识别  标签天线结构参数  改进粒子群算法  多目标寻优

英文关键词:radio frequency identification  tag antenna structure parameters  improved particle swarm optimization algorithm  multi-objective optimization

中文摘要:

      为了解决天线设计人员应用电磁仿真软件优化天线结构时存在的优化方向不明确和优化速度慢的问题,文中 以干式水表的嵌入式射频识别标签天线设计为例,提出了基于改进粒子群算法的标签天线结构参数多目标寻优方法。首 先,根据干式水表产品追溯需求,提出了中心频点尽可能接近理想中心频点、回波损耗尽可能低、带宽尽可能宽、面积尽可 能小的四个目标函数。其次,为避免粒子群算法陷入局部最优,采用多维均匀拉丁超立方初始化、Logistic 混沌映射非线性 变化惯性权重、网格划分变化学习因子、高斯扰动策略等方法对算法进行改进,并应用于标签天线结构参数多目标优化 中。最后,进行了实例验证。验证结果表明:利用改进后的粒子群算法得到的标签天线结构参数优化结果可更大程度满 足优化目标需求,优化耗时仅为电磁仿真软件的40.1%。

英文摘要:

      In order to solve the problems of unclear optimization direction and slow optimization speed in antenna structure optimization by antenna designers using electromagnetic simulation software, the embedded radio frequency identification tag antenna design of dry water meter is taken as an example and a multi-objective optimization method for tag antenna structure parameters based on improved particle swarm optimization (PSO) algorithm is proposed in this paper. Firstly, based on the traceability requirements of dry water meter products, four objective functions are proposed: the center frequency point should be as close as possible to the ideal center frequency point, the return loss should be as low as possible, the bandwidth should be as wide as possible, and the area should be as small as possible. Secondly, in order to avoid the local optima of PSO algorithm, methods such as multidimensional uniform Latin hypercube initialization, logistic chaotic mapping nonlinear change inertia weight, grid partitioning change learning factor, Gaussian perturbation strategy, etc. are used to improve the algorithm and are applied to multi-objective optimization of tag antenna structural parameters. Finally, an instance validation is conducted. The validation results show that the optimized parameters of the tag antenna structure obtained by the improved PSO algorithm meet the target optimization requirements to a greater extent, and the optimization time is only 40. 1% of that of the electromagnetic simulation software.