A micro-gas sensor array was composed of the six tin oxide thin film gas sensors with Au, Cu, or Pt metal catalyst. The sensing properties of sensors to the volatile organic compounds (VOCs) were evaluated. VOCs mean VOCsmixture and formaldehyde (HCHO) in the paper, while the VOCsmixture is the mixtures of 10 ppm toluene, 1 ppm acetone, 5 ppm α-pinene and 10 ppm ethanol. And a multilayer neural network with an error backpropagation (BP) learning algorithm was implemented to recognize the low concentration VOCsmixture and HCHO. The results demonstrated that the micro-sensor array, plus multilayer neural network, was very effective to recognize not only low concentration single component VOCsmixture and HCHO, but also HCHO and VOCsmixture in binary mixture with small errors.%六个由贵金属Au,Cu,Pt做添加剂的SnO2气体传感器构成了微气体传感器阵列.首先研究了这六只传感器对挥发性有机化合物(VOCs)敏感特性,本文中的VOCs 指VOCsmixture和甲醛(HCHO)气体,其中VOCsmixture是10 ppm甲苯、1 ppm丙酮、5 ppm α-派烯和10 ppm乙醇的混合气.然后采用BP神经网络对所获得的传感器信号进行了分析、识别.结果显示微气体传感器阵列与BP神经网络相结合不仅能有效地识别低浓度的单成分VOCsmixture和甲醛气体,而且也能有效地识别两元气体中的VOCsmixture和甲醛气体.
展开▼
