基于传感器阵列与神经网络的气体检测系统

在分析研究电子鼻理论和系统组成的基础上,设计构建了一套传感器阵列与人工神经网络相结合的混合气体检测系统.并采用该系统对三种气体传感器(一氧化碳CO、二氧化硫SO2和二氧化氮NO2)进行了实验,对实验数据用神经网络(BP和 RBF)进行了分析、识别和气体体积分数的计算.结果显示该检测系统识别准确,不仅能够解决气体传感器交叉敏感问题,提高器件的选择性,而且具有智能化和多功能化等优点.

A Gas Testing System Based on the Gas Sensor Array and the Neural Network

Based on the study of the theory and constituent of the electronic nose system, a set of combined gas sensor array system with artificial neural network, for detection of gas mixture is designed and constructed. Three gas sensors(CO,SO 2,NO 2) are experimented by the system, and the data are analyzed,identified by using artificial neural network(BP: back propagation and RBF: radial basis function), from which the volume fractions of gases are calculated.The research results show that the identification of the system is precise.It solves the problem of the gas cross sensitivity, helping to improve the gas sensor selective, realize the artificial intelligence and multifunction.