微带径向短截线基于知识的人工神经网络模型  

Knowledge-Based Artificial Neural Network Models for Microstrip Radial Stub

作　　者：李超[1] 薛良金[1] 徐军[1] 

机构地区：[1]电子科技大学应用物理研究所,四川成都610054

出　　处：《电子学报》2001年第12期1696-1698,共3页Acta Electronica Sinica

摘　　要：微带径向短截线具有比直微带短截线在更宽的频率范围内实现低阻抗值的优点 .本文采用基于知识的人工神经网络模型模拟微带径向短截线的特性 ,利用已经具有的先验知识减小神经网络输入输出映射关系的复杂程度有效减少了训练样本的数量 .本文建立的人工神经网络模型不仅保留了全波有限元法的准确性 。Microstrip radial stubs are a superior choice over low characteristic microstrip impedance rectangular stubs in terms of maintaining a low input impedance value over a wide frequency range.In this paper,a knowledge based artificial neural network is used to model the microstrip radial stub.Utilizing prior knowledge for reducing complexity of input output relationships that the ANN must learn,it allows an accurate ANN model to be developed with less training data which is very advantageous when training data is expensive/time consuming to obtain,such as with EM simulation.

关 键 词：神经网络模型 知识 微带径向短截线 

分 类 号：TN811[电子电信—信息与通信工程]