Classification And Identification Of Vinegar Based On Gas Sensor Array

Vinegar is the important liquid of our life.We pay many attention on its quality. Therefore, it relates to consumers' health that liquors are quickly and accurately measured.The electronic nose is the cheaper faster and More accurate than any other testing technology. In this paper, a sensor array composed by ten commercial TGS sensors of Figaro company, and the electronic nose system made in laboratory are adopted to dynamically measure vinegar, classification and recognition of different kinds of liquors are analyzed. We chosed the top Sampling device in this Experiment. According to percentages of vinegars, aromatic types and raw materials, 8 vinegars were recognized by using BP-ANN, PCA, LVQ and CA. We can find a right way to differentiate these 8 kinds vinegar by PCA analysis of Categories, raw materials, acidity, fermentation method.And these signals of E-nose are affected by these vinegars'characteristic.But we can not describe the relation between them exactly now.Two kinds of these vinegars can be divided by CA analysis.They are No 1,2,3,4 and No 5,6,7,8.But the accuracy rating of CA is low.We could explain the reason of the situation now.The result of the LVQ analysis is 87.5%,64.3%,65.6%,and 68%, which was analysed by the features of productive technology,place of production,acidity and staple. The result is not contradict with the PCA. Fermentation is the best feature for the Categories of vinegar in LVQ analysis.Because of the same productive technology,the mature vinegar and aromatic vinegar can not be discriminated by LVQ.And the BP-ANN analysis has a low index of the vingers'identification.Its rate of identification is only 54.17%.The rate of the specimen identification is 100%.therefore,there is 77.085% of the rate of identification in General.