摘要
将气体传感器阵列与神经网络模式识别技术相结合,构建了一个多组分混合气体定量检测系统,进行了初步的试验,并对试验方法进行了改进。结果表明,用该方法识别准确率接近93%,为畜舍内空气质量的在线检测打下了良好的基础。
The gas sensor array was combined with artificial neural network pattern recognition to form a mix-gas quantitative analysis system,Primarily test result of the system and some optimizing methods to the experiment were also given.The result showed that new technique was precise and effective in identification with recogzing precise rate of near 93%, and was feasible for practical multi-gas quantitative analysis system at swine buildings.
出处
《安徽农业科学》
CAS
北大核心
2007年第12期3744-3745,共2页
Journal of Anhui Agricultural Sciences
关键词
畜舍
多组分气体
径向基函数神经网络
传感器阵列
Swine building
Multi component-gas
Radial basis function neural network
Sensor array
