基于傅立叶变换中红外光谱技术的浓香型基酒快速检测OACSTPCD

To monitor the quality of different distillates in the brewing process of strong-flavor Baijiu online,rapid detection models of eight compounds from strong-flavor base Baijiu in the process of picking were established through Fourier transform mid-infrared spectros-copy combined with partial least square method.The eight compounds include ethanol,ethyl acetate,ethyl butyrate,ethyl caproate,ethyl lactate,acetic acid,caproic acid and butyrate.The determination coefficients R2 of the prediction model of alcohol content,ethyl acetate and ethyl lactate were 0.99,that of ethyl caproate was 0.95,that of ethyl butyrate,acetic acid caproic acid are about 0.90,and that of butyrate was 0.79.The predicted results of infrared spectra on eight compounds had good linear correlation with chemical values,and the wider range of compound content,the more uniform content distribution,the better fitting degree of the model.The temperature difference in different seasons caused the fluctuation of infrared absorption spectrum of base Baijiu,which will reduce the accuracy of model detec-tion.The detection model established by the method of temperature compensation global correction can solve this problem and provide a feasible scheme for realizing automatic online Baijiu picking.