基于支持向量回归的腔体滤波器机电耦合建模与优化  

Modeling and Optimization for Electromechanical Coupling of Cavity Filters Based on Support Vector Regression

作　　者：周金柱[1] 黄进[1] 薛欣[2] 

机构地区：[1]西安电子科技大学电子装备结构设计教育部重点实验室,西安710071 [2]西安电子科技大学天线与微波技术国家重点实验室,西安710071

出　　处：《电子与信息学报》2011年第11期2780-2784,共5页Journal of Electronics & Information Technology

基　　金：国家973计划项目(61358);国家自然科学基金(10433020);中央高校基本科研业务费(JY10000904019)资助课题

摘　　要：为了提高装配后的滤波器电性能和成品率,该文提出了一种基于支持向量回归的腔体滤波器机电耦合建模和优化方法。该方法根据工程中积累的经验数据,首先应用改进的多核线性规划支持向量回归算法建立了制造精度对滤波器电性能影响的机电耦合模型,然后应用此模型优化了滤波器的制造精度,从而获得了最优的机械结构尺寸。实际滤波器的实验结果验证了该方法的有效性。该方法可用于批量生产的腔体滤波器的计算机辅助制造系统中。A modeling and optimization method for electromechanical coupling of cavity filters is proposed to improve the electrical performance and yield rate of assembled filters.In the method,a coupling model that can reveal the effect of manufacturing precision on electrical performance of cavity filters is developed by an improved multi-kernel linear programming support vector regression,according to some data from the manufacturing process of filters.Then,the manufacturing precision is optimized by using the coupling model,and the optimal mechanical structure is obtained.Some experiments from a practical filter have been carried out,and the results confirm the effectiveness of the proposed approach.The approach is particularly suitable to the computer-aided manufacturing of volume-producing filters.

关 键 词：腔体滤波器 机电耦合 线性规划支持向量回归 多核 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]