摘要
用金属氧化物半导体气敏传感器阵列组成的人工嗅觉系统 (电子鼻 )对 3种品牌卷烟烟气进行分析 详细阐述了实验过程及确定传感器的组成方法 ,并从样本中用马氏距离选取合适的样本 ,用主成分分析法和神经网络聚类分析法对样本进行分析 主成分分析结果是已较好地把各品牌的烟分开 ,神经网络对 3种品牌烟的识别率分别为玉溪 85%、白沙 90 %、红梅 95%
An electronic nose system composed of an oxide semiconductor gas sensor array is used to analyze cigarettes of three different brands.A detailed exposition is given to the processes of the experiment and the components of the sensor array. The satisfying samples are obtained from initial samples by using Mahalanobis distance method. Principal component analysis (PCA) and SOM(Self Organizing feature Map) neural network recognition analysis are used to identify cigarettes smoke samples from the three different brands. Good separation among the gases with different brand samples is obtained using the principal component analysis. The recognition probability of the neural network is 85% for Yuxi,90%for Baisha and 95% for Hongmei.
出处
《江苏理工大学学报（自然科学版）》
2000年第3期1-4,共4页
Journal of Jiangsu University of Science and Technology(Natural Science)
基金
江苏省应用基础基金资助项目! (BJ970 6 1)
关键词
卷烟烟气
人工嗅觉系统
神经网络
cigarette smoke
electronic nose system
neural network
sensor array
