Identification and Pattern Recognition Analysis of Chinese Liquors by Doped Nano Zno Gas Sensor Array

QY Zhang,CS Xie,SP Zhang,AH Wang,BL Zhu,L Wang,ZB Yang
DOI: https://doi.org/10.1016/j.snb.2005.02.017
2005-01-01
Abstract:In this paper, ZnO nanoparticles were prepared by the renovated hybrid induction and laser heating. The sensitivities of gas sensors can be greatly improved by doping MnO2, TiO2 and Co2O3. Five different Chinese liquors, namely, Baiyunbian, Beijing Erguotou, Red Star Erguotou, Zhijiangdaqu and Jianliliangjiu, alcohol and diluted alcohol (forged liquor) were measured. Though the main ingredient is alcohol in liquors, Chinese liquors mainly differ from their flavour types, which rely on their trace components. Flavour type of liquors is a very important factor in Chinese liquor identification. Principal component analysis incorporating with discriminant analysis (PCA-DA), back-propagation artificial neural network (BP-ANN) and learning vector quantization (LVQ) were compared for their classification ability. The accuracy of PCA-DA, BP-ANN and LVQ in terms of predicting tested samples was 76.8, 71.4 and 89.3%, respectively. The LVQ is the most suitable pattern recognition algorithm in present experiment. This work shows the potential application of the gas sensor arrays for monitoring the quality of Chinese liquors.
What problem does this paper attempt to address?