Food and Fermentation Industries >

2026 , Vol. 52 >Issue 17: 336 - 345

DOI: https://doi.org/10.13995/j.cnki.11-1802/ts.045311

Traceability analysis of Baijiu by near infrared spectroscopy combined with machine learning

  • DU Yanhong ,
  • HU Chunhong ,
  • TAN Hao ,
  • LI Simai ,
  • CHEN Sijia ,
  • LI Xiao ,
  • GUO Lei ,
  • WANG Shengjie ,
  • XU Yan ,
  • TANG Ke
Expand
  • 1(College of Bioengineering, Jiangnan University, Wuxi 214122, China)
    2(Beijing Red Star Co.Ltd., Beijing 101400, China)

Received date: 2025-11-27

  Revised date: 2026-01-21

  Online published: 2026-09-21

Fold

Abstract

Taking the 2025 national Baijiu taster training samples as the experimental object, the gas chromatography, near-infrared spectroscopy technology combined with machine learning were used to analyze the different flavor substances of different flavor types, origins and quality grades of Baijiu, and to construct the discrimination model.The results showed that PLS-DA could effectively distinguish the characteristics of different types of Baijiu flavor substances.Taking VIP value >1, P<0.05 as the standard, 21 kinds of Baijiu with different flavor types were identified as β-phenylethyl alcohol, n-butanol, acetal, etc;27 kinds of Jiangxiangxing Baijiu from different regions were identified as ethyl palmitate, β-phenylethanol, ethyl oleate, etc;21 kinds of volatile flavor compounds such as isobutanol, β-phenylethanol, diethyl succinate, etc;and 14 kinds of volatile flavor compounds such as n-pentanol, methanol and isobutyraldehyde were identified as important contribution to different quality grades of Qingxiangxing Baijiu.Compared with RF and SVM, GA-BP model has excellent discrimination performance for different flavor types, origins and quality grades of Baijiu, and the classification accuracy of test samples is 91.33%, 94.38%, 98.28% and 94.83%, respectively.The SHAP interpretation results showed that the wave numbers of 11 996、7 383、5 952,7 128、5 848、5 311,6 935、6 932、5 319,11 996、5 902、5 824 cm-1 had a higher contribution to the model output, and were determined to be the key spectral points for predicting different types of Baijiu.The near-infrared spectroscopy combined with GA-BP model proposed in this study has a high accuracy in the discrimination of different types of Baijiu, which provides a theoretical reference for the rapid traceability and the scientific evaluation of different types of Baijiu.

Key words： near infrared spectroscopy; gas chromatography; machine learning; flavor type; origin; quality grade

Cite this article

DU Yanhong , HU Chunhong , TAN Hao , LI Simai , CHEN Sijia , LI Xiao , GUO Lei , WANG Shengjie , XU Yan , TANG Ke . Traceability analysis of Baijiu by near infrared spectroscopy combined with machine learning[J]. Food and Fermentation Industries, 2026 , 52(17) : 336 -345 . DOI: 10.13995/j.cnki.11-1802/ts.045311

References

[1] Zheng X W, Han B Z.Baijiu, Chinese liquor:History, classification and manufacture[J].Journal of Ethnic Foods, 2016, 3(1):19-25.
[2] 魏泉增, 范江涛, 刘嘉玲, 等.偏最小二乘方法在不同白酒香型判别分析中的应用[J].中国酿造, 2020, 39(10):183-187.
Wei Q Z, Fan J T, Liu J L, et al.Application of partial least squares in different flavor types of Baijiu discriminant analysis[J].Chinese Brewing, 2020, 39(10):183-187.
[3] Wang W, Chen J Q, Chen J S, et al.Spatial distribution of Chinese baijiu driven by groundwater recharge processes:Insights from hydrochemical analysis[J].Groundwater for Sustainable Development, 2025, 31:101513.
[4] 王怡博, 庹先国, 张贵宇, 等.近红外光谱结合改进的排列组合总体分析方法鉴别浓香型白酒基酒等级[J].食品工业科技, 2026,47(2):347-355.
Wang Y B, Tuo X G, Zhang G Y, et al.Near-infrared spectroscopy combined with improved permutation combination population analysis for identifying base liquor grades of Nongxiangxing Baijiu[J].Science and Technology of Food Industry, 2026,47(2):347-355.
[5] Wang G N, Song X B, Zhu L, et al.A flavoromics strategy for the differentiation of different types of Baijiu according to the non-volatile organic acids[J].Food Chemistry, 2022, 374:131641.
[6] 卓俊纳. 三种香型白酒的ICP-OES和GC-QTOF MS分析及判别研究[D].宜宾:四川轻化工大学, 2022.
Zhuo J N.Analysis and discrimination research by ICPOES and GC-QTOF MS of three kinds of flavor Baijiu[D].Yibin:Sichuan University of Science and Engineering, 2022.
[7] 吴李玲, 裴荣红, 李行, 等.不同产地酱香型白酒中挥发性含氮化合物的差异分析[J].食品科学技术学报, 2024, 42(3):35-48.
Wu L L, Pei R H, Li X, et al.Difference analysis of volatile nitroge-containing compounds in soy sauce flavor-type baijiu from different regions in China[J].Journal of Food Science and Technology, 2024, 42(3):35-48.
[8] 夏于林, 尹希杰, 薛康, 等.电感耦合等离子体质谱联合气相色谱-同位素比值质谱法鉴别白酒产地[J].化学分析计量, 2025, 34(5):55-60.
Xia Y L, Yin X J, Xue K, et al.Origin identification of Baijiu based on ICP-MS and GC-IRMS[J].Chemical Analysis and Metrology, 2025, 34(5):55-60.
[9] Li S, Li T, Han Y R, et al.Machine learning discrimination and prediction of different quality grades of sauce-flavor baijiu based on biomarker and key flavor compounds screening[J].Food Chemistry:X, 2024, 24:101877.
[10] 王凡, 张文娟, 李国辉, 等.近红外光谱技术同时快速定量酱香型白酒基酒醇类物质[J].食品与发酵工业, 2025, 51(16):324-331.
Wang F, Zhang W J, Li G H, et al.Simultaneous and rapid quantification of alcohols in Jiangxiangxing Baijiu base liquor by near-infrared spectroscopy[J].Food and Fermentation Industries, 2025, 51(16):324-331.
[11] 黄清霞, 付国勇, 陈黎萍.成品白酒中酸酯总量的近红外检测[J].食品工业科技, 2022, 43(5):310-314.
Huang Q X, Fu G Y, Chen L P.Detection of acid ester content in Baijiu (Chinese liquor) based on near infrared spectroscopy[J].Science and Technology of Food Industry, 2022, 43(5):310-314.
[12] Yang L, Xian C, Li S, et al.Machine learning combined with GC-FID for discrimination of different categories of Maotai-flavor baijiu[J].Food Chemistry:X, 2025, 28:102555.
[13] Xie J, Hong J X, Zhang C S, et al.Machine learning-assisted identification of core flavor compounds and prediction of core microorganisms in fermentation grains and pit mud during the fermentation process of strong-flavor Baijiu[J].Food Chemistry, 2025, 495(2):146426.
[14] 杜艳红, 谭昊, 聂建光, 等.运用多工艺融合技术提升清香型白酒产品品质的研究[J].酿酒科技, 2022(10):65-70.
Du Y H, Tan H, Nie J G, et al.Improvement of Qingxiang Baijiu quality by process fusion[J].Liquor-Making Science & Technology, 2022(10):65-70.
[15] Kasemsumran S, Boondaeng A, Ngowsuwan K, et al.Simultaneous monitoring of the evolution of chemical parameters in the fermentation process of pineapple fruit wine using the liquid probe for near-Infrared coupled with chemometrics[J].Foods, 2022, 11(3):377.
[16] Zhou Q, Huang W Q, Tian X.Feature wavelength selection based on the combination of image and spectrum for aflatoxin B1 concentration classification in single maize kernels[J].Agriculture, 2022, 12(3):385.
[17] 曹梦德, 王怡博, 张贵宇, 等.基于机器学习混合特征选择方法的近红外光谱检测浓香型白酒基酒等级[J].食品与发酵工业, 2025, 51(24):359-367.
Cao M D, Wang Y B, Zhang G Y, et al.Detection of base liquor grade of Nongxiangxing Baijiu by near-infrared spectroscopy based on machine learning hybrid feature selection method[J].Food and Fermentation Industries, 2025, 51(24):359-367.
[18] Yu F H, Feng S, Du W, et al.A Study of nitrogen deficiency inversion in rice leaves based on the hyperspectral reflectance differential[J].Frontiers in Plant Science, 2020, 11:573272.
[19] Zhang L, Schultz M A, Cash R, et al.Determination of quality parameters of tomato paste using guided microwave spectroscopy[J].Food Control, 2014, 40:214-223.
[20] Ma L Y, Chen L, Chou K C, et al.Campylobacter jejuni antimicrobial resistance profiles and mechanisms determined using a raman spectroscopy-based metabolomic approach[J].Applied and Environmental Microbiology, 2021, 87(12):e00388-21.
[21] Wei L F, Zhang Y X, Yuan Z R, et al.Development of visible/near-infrared hyperspectral imaging for the prediction of total arsenic concentration in soil[J].Applied Sciences, 2020, 10(8):2941.
[22] 王凡, 陈鹏, 庞文虎, 等.衰减全反射傅里叶变换红外光谱快速定量酱香型白酒骨架成分[J].中国酿造, 2025, 44(10):281-288.
Wang F, Chen P, Pang W H, et al.Rapid quantification of skeleton compounds of sauce-flavor Baijiu by attenuated total reflection Fourier transform infrared spectroscopy[J].China Brewing, 2025, 44(10):281-288.
[23] 何菲, 蒋洪久, 玉光惠, 等.不同香型白酒香味物质特征差异比较分析[J].酿酒科技, 2022(8):48-53;58.
He F, Jiang H J, Yu G H, et al.Comparative analysis of aroma substance characteristics of baijiu of different flavor types[J].Liquor-Making Science & Technology, 2022(8):48-53;58.
[24] 先春, 陈仁远, 赵文武, 等.感官排序结合多元统计分析对酱香型白酒轮次基酒质量等级的研究[J].中国酿造, 2025, 44(8):170-175.
Xian C, Chen R Y, Zhao W W, et al.Quality grade of round base liquor of sauce-flavor Baijiu by sensory ranking and multivariate statistical analysis[J].China Brewing, 2025, 44(8):170-175.
[25] 张磊, 庹先国, 张贵宇.基酒等级的近红外快速检测和基于SHAP的模型解释[J].中国酿造, 2025, 44(7):283-288.
Zhang L, Tuo X G, Zhang G Y.Near infrared rapid detection of base liquor grades and model interpretation based on SHAP[J].China Brewing, 2025, 44(7):283-288.
[26] 夏珍珍, 张博源, 郑丹, 等.基于PLS与CNN的甘薯淀粉掺假鉴别及量化比较[J].食品科学, 2025, 46(20):327-336.
Xia Z Z, Zhang B Y, Zheng D, et al.Identification and quantification of adulterated sweet potato starch based on PLS and CNN:A comparative study[J].Food Science, 2025, 46(20):327-336.
[27] 余松柏, 黄张君, 赵兴蓉, 等.红外光谱技术及其在白酒酿造过程中快速检测与分析的研究进展[J].分析测试学报, 2025, 44(3):544-561.
Yu S B, Huang Z J, Zhao X R, et al.Research progress on infrared spectroscopy and its rapid detection and analysis in baijiu brewing[J].Journal of Instrumental Analysis, 2025, 44(3):544-561.
[28] 谢丹, 吴成, 赵雯宇, 等.酱香型白酒酒醅中产香白地霉的分离鉴定、特性解析及全基因组分析[J/OL].食品工业科技, 2025.DOI:10.13386/j.issn1002-0306.2025070261.
Xie D, Wu C, Zhao W Y, et al.Isolation, identification and characteristic analysis of aroma-producing Geotrichum candidum from fermented grains of Jiangxiang Baijiu[J/OL].Science and Technology of Food Industry, 2025.DOI:10.13386/j.issn1002-0306.2025070261.
[29] 王允中, 罗安来, 王淑军.多元统计分析方法在白酒香型识别中的应用[J].食品安全导刊, 2021(9):118-122.
Wang Y Z, Luo A L, Wang S J.Application of multivariate statistical analysis methods in baijiu aroma identification[J].China Food Safety Magazine, 2021(9):118-122.
[30] 朱开宪, 胡雪, 邓静, 等.基于GC-MS技术对不同香型白酒的判别分析[J].中国酿造, 2023, 42(1):213-218.
Zhu K X, Hu X, Deng J, et al.Discriminant analysis of different flavor-type Baijiu based on GC-MS[J].China Brewing, 2023, 42(1):213-218.
[31] 向玲, 贾俊杰, 毛雪婷, 等.基于表面增强拉曼光谱技术对不同产地酱香型白酒的鉴别研究[J].分析测试学报, 2026,45(1):77-87.
Xiang L, Jia J J, Mao X T, et al.Study on the identification of sauce-flavor Baijiu from different origins based on surface-enhanced Raman spectroscopy technology[J].Journal of Instrumental Analysis, 2026,45(1):77-87.
[32] Zhou C L, Yu Y S, Zhu Y W, et al.Machine learning discrimination of quality grades of base liquor integrating GC-TOF/MS and GC-IMS data analysis:Case study of strong-flavor Chinese baijiu[J].Food Chemistry, 2025, 492(1):145281.
[33] Su Y Y, Yang Y, Kilmartin P A, et al.Inter-regional characterization of New Zealand pinot noir wines:Assessing geographical origin through mid-FTIR and phenolic profile analysis[J].Food Research International, 2025, 212:116485.
Options
Outlines

/

〈 〉

Powered by Beijing Magtech Co. Ltd,.