【摘要】 采用6个不同掺杂的纳米ZnO气体传感器组成的阵列实现了乙醇、丙酮、苯、甲苯、二甲苯的识别。研究表明,掺杂可大幅度提高传感器的敏感度和对可挥发有机物(VOCs)的选择性。对比了k近邻法、线性判别法、反传人工神经网络、概率神经网络、学习向量量化等在本实验中的应用。反传人工神经网络具有最高识别率,可达100%。本研究表明电子鼻在空气质量监测中具有广阔的应用前景。更多还原

【Abstract】 Recognition of ethanol, acetone, benzene, toluene and xylene was performed by using 6 doped nano ZnO gas sensors. It was proved that sensitivities and selectivity of gas sensors could be reasonably improved by dopants. K-nearest neighbour (k-NN), linear discriminant analysis (LDA), back-propagation artificial neural network (BP-ANN), probabilistic neural network (PNN) and learning vector quantization (LVQ) were compared for their suitability on classifying volatile organic compounds (VOCs). The accuracy of BP-ANN in terms of predicting tested samples was 100% and the highest among the pattern recognition algorithms. This work shows the potential application of the gas sensor arrays for monitoring the air quality.更多还原