| Chinese liquor is a world-famous beverage with a long history for its unique aroma.Base liquor,a product of liquor brewing,significantly affects the flavor and quality of commercial liquors.In this study,six types of base liquors and three types of commercial liquors are selected as research object,E-nose and Gas Chromatography-Mass Spectrometer(GC-MS)are used to detect the characteristic aroma of liquors.In the quality detection of liquor based on E-nose,a machine learning method consisting of a deep residual network Res Net backbone with a light gradient boosting machine(Light GBM)classifier(Res Net-GBM)is proposed to classify the quality of base liquor and commercial liquor using multi-channel signals from an electronic nose(E-nose).The superiority and generalization of the proposed model were demonstrated through ablation experiments and comparative experiments.In the detection of liquor components on GC-MS,91 substances were detected in 9 samples,of which esters had the highest relative content with more than 50%.The proportion of each component in liquors can be understood more intuitively through clustering heatmap.The differences and correlations between different samples were visually demonstrated through Principal Component Analysis(PCA).The analysis results of Partial Least Squares Discriminant Analysis(PLS-DA)provide the types of compounds that contribute the most to the difference of nine kinds of liquors were obtained.In a word,E-nose and GC-MS can realize the qualitative and quantitative detection of liquor’s characteristic aroma.This paper has carried out classification research of characteristic aroma data based on E-nose,while GCMS analyses the characteristic aroma data classified by E-nose from the perspective of analytical chemistry,thus providing a theoretical basis for the research and development of specific sensors and electronic nose.Based on the two kinds of equipment,a more comprehensive study of the samples has been realized,which is of great significance for the quality control and improvement of liquor. |