基于PCA和LVQ混合神经网络算法的电子鼻系统  

An electronic nose system based on the algorithm of PCA and LVQ hybrid neural network

作　　者：文政颖[1] 米捷[1] 

机构地区：[1]河南工程学院计算机科学与工程系,河南郑州450007

出　　处：《电子技术应用》2013年第10期76-79,共4页Application of Electronic Technique

基　　金：国家自然科学基金资助项目(61142010)

摘　　要：为了对食物品质进行非接触式评价,采用6种费加罗金属氧化物半导体传感器阵列设计并研制了可对被测食物进行无损检测的电子鼻系统。系统主要由采样模块、控制模块和上位机组成,并采用主成分分析(PCA)和学习矢量量化(LVQ)混合神经网络模式识别算法对气体"指纹信息"数据库进行分析。实验结果表明,利用该电子鼻系统可以对5种不同的食用酱进行检测,并且具有对未知酱品进行识别的功能。In order to evaluate the food quality without contacting,an electronic nose system composed of acquisition model,controller model and host computer is designed and developed based on six kinds of Figaro metal-oxide semiconductor sensor array,which can test the food without destructing.The Principal Component Analysis （PCA） and Learning Vector Quantization （LVQ） hybrid neural network pattern recognition algorithms are utilized in the proposed system to analyze the fingerprint database of gas.The experimental results show that this instrument can detect five different types of sauces accurately and is capable of distinguishing unknown sauce.

关 键 词：传感器阵列 主成分分析 学习矢量量化 混合神经网络 电子鼻 

分 类 号：TP271[自动化与计算机技术—检测技术与自动化装置]