结合BP神经网络和多目标粒子群算法的圆极化天线优化方法  

Design of multi-objective antenna based on neural network

作　　者：张子洋 储龙威 胡海荣 赵国华 ZHANG Ziyang;CHU Longwei;HU Hairong;ZHAO Guohua(Faculty of Electrical Engineering and Computer Science,Ningbo University,Ningbo 315211,China)

机构地区：[1]宁波大学信息科学与工程学院,宁波315211

出　　处：《无线通信技术》2025年第1期19-22,26,共5页Wireless Communication Technology

基　　金：国家自然科学基金面上项目(62371266);宁波市自然科学基金(2022J097);国家自然科学基金(62271273)。

摘　　要：为了提高天线优化的效率,本文提出了一种结合误差反向传播网络(BPNN)和多目标粒子群算法(MOPSO)的方法来优化多目标天线。通过BPNN模型来拟合天线参数和天线性能之间的非线性函数关系,将训练好的BPNN作为MOPSO的目标函数进行寻优,得到帕累托解集,与之相对应的即为符合要求的天线尺寸。本文使用一个工作频率在1.5GHz-3.4GHz的圆极化天线来验证该方法的有效性,将天线的S11和AR作为优化目标。验证结果表明,通过对天线参数的优化,天线的-10dB带宽和3dB带宽均得到了扩展,优化成效显著。该算法可以实现对天线的优化,相较于传统的人工反复调试方法,节省了大量的计算资源,并且避免了为了优化一个子目标而使得另外一个子目标被负优化的情况出现。This article proposes a method based on BPNN(Backpropagation Neural Network)and MOPSO(Multi-Objective Particle Swarm Optimization)algorithm for optimization multi-objective antennas.The BPNN model is used to fit the nonlinear function relationship between antenna parameters and antenna performance.The trained BPNN is then employed as the objective function in MOPSO for optimization,resulting in a Pareto solution set where corresponding antenna dimensions meet the requirements.To validate the effectiveness of this method,a circularly polarized antenna operating at 1.5GHz-3.4GHz frequency is used,with S11 and AR of the antenna as optimization objectives.Experimental results demonstrate that this algorithm can achieve antenna optimization.Compared to traditional manual iterative methods,it saves a significant amount of computational resources and avoids situations where optimizing one objective negatively impacts another subobjective.

关 键 词：神经网络 多目标优化 帕累托解集 圆极化 

分 类 号：TN821[电子电信—信息与通信工程]