The brand and purity of orange juice and liquor were tested by electronic tongue method

Imitating the mechanism of human taste perception,an electronic tongue system for beverage quality based on virtual instrument and internet of things was developed. During the detection,the "fingerprint" data of orange juice and liquor were first acquired by sensor array which was stimulated by a large pulse scan,then the feature information was extracted through discrete wavelet transform(DWT)from the feedback data of sensor array,after that,the principal component analysis(PCA)and partial least squares regression analysis(PLSR)were applied for the qualitative and quantitative analysis of these information. The analysis results were uploaded to the server through the internet of things,and in which the users could access it by the mobile phones or other terminals. The developed electronic tongue system was further used to the orange juice and liquor brand distinction and purity detection. The results showed that the identification accuracy for different brand of orange juice and liquor reached 100%,while the correlation coefficient R² in the validation set were 0.985 and 0.998,as well as the relative percent deviations RPD in validation set were 2.895 and 3.272.This system possessed the advantages of fast detection speed,simple operation,low cost,stability and reliability as well as the result was easy to query.