Optimization Method of Gas Sensor Array for Identification of Jing Wine Based on Electronic Nose

Optimization of gas sensor array is an important topic in the application of electronic nose. The identification ability of electronic nose was improved and the redundant information was reduced by array optimization. Three methods were used to optimize the original array in the identification of Jing Wine respectively, they are the minimum cumulative of absolute correlation coefficient, the second principal component coefficients selective analysis and the analysis of the load factor based on the coefficient of variation. The results showed that three kinds of optimization array forms corresponding to the three methods were about the same, and then a sensor array that could be met the discrimination for Jing Wine of 38 and 35 vol% alcohol was constructed. Computing results indicate that optimization sensor array could make better on identification of the two kinds of alcohol concentration. Therefore a method and idea of sensor array optimization for drink discrimination was also put forward as a result of this investigation.