为探究无损、快速和准确判别不同类型浓香型白酒的方法,该研究选取不同酒度和不同品牌浓香型白酒作为研究对象。利用傅里叶变换中红外光谱仪采集120份酒样的原始光谱,结合平滑滤波与标准正态变换分别对原始光谱进行预处理,采用主成分分析比较光谱预处理效果。光谱数据按Kennard-Stone方法以7∶3的比例划分为训练集和测试集,经数据归一化后使用蚱蜢算法优化支持向量机和误差反向传播人工神经网络进行建模分析。试验结果表明,光谱预处理结合主成分分析不能区分不同酒度和品牌的浓香型白酒,但平滑滤波处理后不同酒度酒样的聚类区分较好,标准正态变换处理后不同品牌酒样的聚类区分更好,二者都能有效减少中红外光谱的噪音,提高识别的精度。基于蚱蜢算法优化支持向量机和误差反向传播人工神经网络模型进行判别时,训练集和测试集的酒样分类准确率均为100%。综上所述,利用中红外光谱结合化学计量学可快速准确地判别不同酒度以及不同品牌浓香型白酒,可为白酒的香型区分、产地溯源、市场监管和售后管理等提供数字化方案。
周瑞
,
陈晓明
,
张莉丽
,
张良
,
许德富
,
张宿义
,
代小雪
,
毛洪川
,
谢菲
,
代汉聪
,
宋艳
,
郭佳
,
陈雯月
. 化学计量学结合中红外光谱的浓香型白酒分类研究[J]. 食品与发酵工业, 2023
, 49(5)
: 88
-93
.
DOI: 10.13995/j.cnki.11-1802/ts.031674
In order to explore the non-destructive, rapid and accurate method of distinguishing different types of strong-flavor Baijiu, different alcoholic strength and different brands of strong-flavor Baijiu were selected as the research object in this study. Using the Fourier transform mid-infrared spectrometer to collect the original spectra of 120 Baijiu samples, combining the smoothing filtering and the standard normal variate method to preprocess the original spectra respectively, and the principal component analysis was used to compare the spectral preprocessing effects. The spectral data were divided into training set and test set according to the Kennard-Stone method with a ratio of 7∶3. After the data normalized, the grasshopper algorithm was used to optimize the support vector machine and the error back-propagation artificial neural network for modeling and analysis. The test results showed that spectral preprocessing combined with principal component analysis cannot distinguish strong-flavor Baijiu with different alcoholic strength and brands, but the clustering distinction of Baijiu samples with different alcoholic strength after smoothing filtering treatment was better, and the clustering distinction of different brands of Baijiu samples after standard normal variate processing was better, both of them can effectively reduce the noise of mid-infrared spectrum and improve the recognition accuracy. When the discriminant analysis was performed based on the grasshopper algorithm was used to optimize the support vector machine and the error back-propagation artificial neural network models, the classification accuracy of Baijiu samples in both the training set and the test set was 100%. In summary, the method of mid-infrared spectroscopy combined with chemometrics can identify strong-flavor Baijiu with different alcoholic strength and brands quickly and accurately, and can provide digital solutions for Baijiu aroma differentiation, origin traceability, market supervision and after-sales management.
[1] 毛洪川. 整粒高粱酿造浓香型白酒的工艺开发及应用研究[D].绵阳:西南科技大学, 2017.
MAO H C.Development and application of a new technology of Luzhou flavour liquor using intact sorghum as raw material[D].Mianyang:Southwest University of Science and Technology, 2017.
[2] 胡耀强, 郭敏, 叶秀深, 等.近红外光谱法间接测定白酒酒精度[J].光谱学与光谱分析, 2022, 42(2):410-414.
HU Y Q, GUO M, YE X S, et al.Indirect determination of liquor alcohol content based on near-infrared spectrophotometry[J].Spectroscopy and Spectral Analysis, 2022, 42(2):410-414.
[3] 刘建学, 杨国迪, 韩四海, 等.白酒基酒中典型醇的近红外预测模型构建[J].食品科学, 2018, 39(2):281-286.
LIU J X, YANG G D, HAN S H, et al.Prediction model for typical alcohols in base liquor based on near infrared spectroscopy[J].Food Science, 2018, 39(2):281-286.
[4] YANG B, YAO L J, PAN T.Near-infrared spectroscopy combined with partial least squares discriminant analysis applied to identification of liquor brands[J].Engineering, 2017, 9(2):181-189.
[5] 周军, 杨洋, 姚瑶, 等.中红外光谱技术在浓香型白酒原酒关键指标分析中的应用[J].光谱学与光谱分析, 2022, 42(3):764-768.
ZHOU J, YANG Y, YAO Y, et al.Application of mid-infrared spectroscopy in the analysis of key indexes of strong flavour Chinese spirits base liquor[J].Spectroscopy and Spectral Analysis, 2022, 42(3):764-768.
[6] XIAO S J, WANG Q H, LI C F, et al.Rapid identification of A1 and A2 milk based on the combination of mid-infrared spectroscopy and chemometrics[J].Food Control, 2022, 134:108659.
[7] PEI Y F, ZUO Z T, ZHANG Q Z, et al.Data fusion of Fourier transform mid-infrared (MIR) and near-infrared (NIR) spectroscopies to identify geographical origin of wild Paris polyphylla var.yunnanensis[J].Molecules (Basel, Switzerland), 2019, 24(14):2559.
[8] 第五鹏瑶. 光谱预处理方法及其集成研究[D].天津:天津工业大学,2019.
DIWU P Y.Spectral preprocessing method and its integration research[D].Tianjin:Tianjin Polytechnic University, 2019.
[9] 田琼, 马新华, 袁俊杰, 等.基于主成分分析和人工神经网络的近红外光谱大豆产地识别[J].食品工业科技, 2021, 42(9):270-274.
TIAN Q, MA X H, YUAN J J, et al.Soybean origin identification based by near-infrared spectrum based on principal component analysis and artificial neural network model[J].Science and Technology of Food Industry, 2021, 42(9):270-274.
[10] SAREMI S, MIRJALILI S, LEWIS A.Grasshopper optimisation algorithm:Theory and application[J].Advances in Engineering Software, 2017, 105:30-47.
[11] 吕赵明, 张颖江.基于改进GOA-SVM算法的异常流量识别[J].湖南科技大学学报(自然科学版), 2019, 34(4):90-96.
LV Z M, ZHANG Y J.Abnormal traffic identification based on improved GOA-SVM algorithm[J].Journal of Hunan University of Science ＆ Technology (Natural Science Edition), 2019, 34(4):90-96.
[12] BLAZHKO U, SHAPAVAL V, KOVALEV V, et al.Comparison of augmentation and pre-processing for deep learning and chemometric classification of infrared spectra[J].Chemometrics and Intelligent Laboratory Systems, 2021, 215:104367.
[13] ABDULLAH S, PRADHAN R C, PRADHAN D, et al.Modeling and optimization of pectinase-assisted low-temperature extraction of cashew apple juice using artificial neural network coupled with genetic algorithm[J].Food Chemistry, 2021, 339:127862.
[14] MA Q X, TENG Y, LI C, et al.Simultaneous quantitative determination of low-concentration ternary pesticide mixtures in wheat flour based on terahertz spectroscopy and BPNN[J].Food Chemistry, 2022, 377:132030.
[15] GU P, FENG Y Z, ZHU L, et al.Unified classification of bacterial colonies on different agar media based on hyperspectral imaging and machine learning[J].Molecules (Basel, Switzerland), 2020, 25(8):1797.
