食品与发酵工业 >

2025 , Vol. 51 >Issue 24: 368 - 374

DOI: https://doi.org/10.13995/j.cnki.11-1802/ts.043167

分析与检测

基于异常值双重剔除与改进偏最小二乘法算法的近红外光谱红缨子糯高粱关键指标快速检测模型优化研究

  • 肖徐 ,
  • 袁进 ,
  • 李静 ,
  • 王薇娜 ,
  • 聂叶 ,
  • 焦富 ,
  • 杨洁
展开
  • (贵州茅台酒股份有限公司,贵州 仁怀,564501)
第一作者:硕士,助理工程师(杨洁助理工程师为通信作者,E-mail:Yangie_Meave@163.com)

收稿日期: 2025-04-29

  修回日期: 2025-06-16

  网络出版日期: 2026-01-12

收起

Optimization of a rapid test model for key indicators of glutinous sorghum cultivar “Hongyingzi” in near-infrared spectroscopy based on outlier double rejection and improved partial least squares algorithm

  • XIAO Xu ,
  • YUAN Jin ,
  • LI Jing ,
  • WANG Weina ,
  • NIE Ye ,
  • JIAO Fu ,
  • YANG Jie
Expand
  • (Guizhou Moutai Wine Co.Ltd., Renhuai 564501, China)

Received date: 2025-04-29

  Revised date: 2025-06-16

  Online published: 2026-01-12

Fold

摘要

红缨子糯高粱是酱香型白酒的核心酿酒原料,检测其关键指标可保障白酒品质,但现有检测方法耗时长,因此,该研究旨在开发一种高效红缨子糯高粱关键指标检测方法。以手工检测值为参考,采用近红外光谱技术采集样品光谱,分别使用箱线图四分位距(interquartile range,IQR)、主成分分析(principal component analysis,PCA)和Hotelling T2检验剔除异常检测值与光谱数据,用SPXY(sample set partitioning based on joint X-Y distances)法划分样本校正集与验证集,通过改进偏最小二乘法(partial least squares,PLS)构建了高粱水分和淀粉含量的快速检测模型。所建模型的定标相关系数(R-squared,RSQ)均超过0.80,标准偏差(square error of calibration,SEC)分别为0.11和0.47,交互验证标准偏差(standard error of cross validation,SECV)分别为0.11和0.59,交互验证相关系数(1 minus the variance ratio, 1-VR)分别为0.81和0.79,表明模型具有良好的线性关系。进一步验证后发现,模型预测与手工检测结果偏差小,表明模型准确性和重复性较高、预测性能优异,研究结果为红缨子糯高粱的快速检测提供了可靠的工具,具有实际生产应用价值。

关键词： 近红外光谱; 红缨子糯高粱; Hotelling T2检验; SPXY算法; 关键指标

本文引用格式

肖徐 , 袁进 , 李静 , 王薇娜 , 聂叶 , 焦富 , 杨洁 . 基于异常值双重剔除与改进偏最小二乘法算法的近红外光谱红缨子糯高粱关键指标快速检测模型优化研究[J]. 食品与发酵工业, 2025 , 51(24) : 368 -374 . DOI: 10.13995/j.cnki.11-1802/ts.043167

Abstract

Glutinous sorghum cultivar “Hongyingzi” is a core raw material for Jiang-flavor Baijiu, but traditional methods for testing its key indicators are time-consuming.This study aimed to develop an efficient near-infrared spectroscopy (NIRS) method.Using manual test values as references, spectral data were collected, and outliers were removed via boxplot IQR analysis and Hotelling's T2 test after PCA.Samples were divided into calibration and validation sets using the SPXY algorithm.Modified partial least squares (PLS) was employed to construct rapid detection models for moisture and starch content.The calibration models showed determination coefficients (R2) above 0.80, with standard errors of calibration (SEC) of 0.11 and 0.47, and cross-validation errors (SECV) of 0.11 and 0.59, respectively.The 1 minus the variance ratio(1-VR) reached 0.81 and 0.79, indicating strong linearity.Random sampling validation confirmed minimal deviations from traditional methods, demonstrating high accuracy and repeatability of the constructed models.The proposed models provide a reliable tool for rapid quality control of glutinous sorghum cultivar “Hongyingzi” in industrial applications.

Key words： near infrared spectroscopy; glutinous sorghum cultivar “Hongyingzi”; Hotelling T2 test; SPXY algorithm; key indicators

参考文献

[1] 倪德让, 孔祥礼, 孙崇德, 等.红缨子高粱淀粉分子结构及糊化特性研究[J].中国酿造, 2019, 38(12):75-79.
NI D R, KONG X L, SUN C D, et al.Molecular structure and gelatinization characteristics of starch from Hongyingzi sorghum[J].China Brewing, 2019, 38(12):75-79.
[2] 陈定崑, 李巧玉, 胡宇佳, 等.近红外光谱技术在茅台酒酒醅检测中的应用[J].酿酒科技, 2021(1):55-58.
CHEN D K, LI Q Y, HU Y J, et al.Application of near infrared spectroscopy in the detection of fermented grains of maotai liquor[J].Liquor-Making Science & Technology, 2021(1):55-58.
[3] 杜伯会, 滕江波, 邹惠玲.现代近红外光谱分析在食品检测中的应用[J].中国食品工业, 2022(22):30-31.
DU B H, TENG J B, ZOU H L.Application of modern near-infrared spectral analysis in food testing[J].China Food Industry, 2022(22):30-31.
[4] 袁颉, 毛育志, 姚翠萍, 等.近红外光谱技术在酱香型大曲糖化力快速检测中的应用[J].酿酒科技, 2023(12):110-113.
YUAN J, MAO Y Z, YAO C P, et al.Application of near infrared spectroscopy in rapid detection of saccharification power of Jiangxiang Daqu[J].Liquor-Making Science & Technology, 2023(12):110-113.
[5] 余松柏, 黄张君, 吴奇霄, 等.基于近红外光谱构建酒用高粱主要理化指标的快速无损分析模型[J].食品工业科技, 2023, 44(10):311-319.
YU S B, HUANG Z J, WU Q X, et al.Constructing rapid and undamaged detection models for main physicochemical indexes of brewing sorghum based on near infrared spectrum[J].Science and Technology of Food Industry, 2023, 44(10):311-319.
[6] 苏鹏飞, 刘丽丽, 闫宗科, 等.基于近红外高粱水分的快速分析研究[J].酿酒, 2021,48(1):87-90.
SU P F, LIU L L, YAN Z K, et al.Study on NIR rapid analysis of water content in the sorghum[J].Liquor Making, 2021, 48(1):87-90.
[7] 苏鹏飞, 张武岗.基于NIR高粱淀粉含量快检技术的开发研究[J].酿酒科技, 2022(2):107-110.
SU P F, ZHANG W G.Development of rapid detection technology for starch in sorghum based on NIR[J].Liquor-Making Science & Technology, 2022(2):107-110.
[8] 张北举, 陈松树, 李魁印, 等.基于近红外光谱的高粱籽粒直链淀粉、支链淀粉含量检测模型的构建与应用[J].中国农业科学, 2022, 55(1):26-35.
ZHANG B J, CHEN S S, LI K Y, et al.Construction and application of detection model for amylose and amylopectin content in sorghum grains based on near infrared spectroscopy[J].Scientia Agricultura Sinica, 2022, 55(1):26-35.
[9] CEN H Y, HE Y.Theory and application of near infrared reflectance spectroscopy in determination of food quality[J].Trends in Food Science & Technology, 2007, 18(2):72-83.
[10] CHEN Q S, LIN H, ZHAO J W.Near-Infrared Spectroscopy Technology in Food[M].Singapore:Springer Singapore, 2021:23-58.
[11] 褚小立, 史云颖, 陈瀑, 等.近五年我国近红外光谱分析技术研究与应用进展[J].分析测试学报, 2019, 38(5):603-611.
CHU X L, SHI Y Y, CHEN P, et al.Research and application progresses of near infrared spectroscopy analytical technique in China in past five years[J].Journal of Instrumental Analysis, 2019, 38(5):603-611.
[12] SCHWERTMAN N C, OWENS M A, ADNAN R.A simple more general boxplot method for identifying outliers[J].Computational Statistics & Data Analysis, 2004, 47(1):165-174.
[13] WILLIAMS J D, WOODALL W H, BIRCH J B, et al.Distribution of Hotelling's T2 statistic based on the successive differences estimator[J].Journal of Quality Technology, 2006, 38(3):217-229.
[14] APINANTANAKON W, SUNAT K, KINMOND J A.Optimal data division for empowering artificial neural network models employing a modified M-SPXY algorithm[J].Engineering & Applied Science Research, 2019, 46(4):276-284.
[15] WANG G, YIN S.Quality-related fault detection approach based on orthogonal signal correction and modified PLS[J].IEEE Transactions on Industrial Informatics, 2015, 11(2):398-405.
[16] 蒙秋霞, 牛宇, 张丽珍, 等.近红外透射光谱技术测定黍稷蛋白含量的研究[J].中国农学通报, 2011, 27(33):243-247.
MENG Q X, NIU Y, ZHANG L Z, et al.Analysis of protein contents in Panicum miliaceum L.by near infrared transmittance spectroscopy[J].Chinese Agricultural Science Bulletin, 2011, 27(33):243-247.
[17] FERNÁNDEZ PIERNA J A, WAHL F, DE NOORD O E, et al.Methods for outlier detection in prediction[J].Chemometrics and Intelligent Laboratory Systems, 2002, 63(1):27-39.
[18] 刘蓉, 陈文亮, 徐可欣, 等.奇异点快速检测在牛奶成分近红外光谱测量中的应用[J].光谱学与光谱分析, 2005, 25(2):207-210.
LIU R, CHEN W L, XU K X, et al.Fast outlier detection for milk near-infrared spectroscopy analysis[J].Spectroscopy and Spectral Analysis, 2005, 25(2):207-210.
[19] 闵顺耕, 李宁, 张明祥.近红外光谱分析中异常值的判别与定量模型优化[J].光谱学与光谱分析, 2004, 24(10):1205-1209.
MIN S G, LI N, ZHANG M X.Outlier diagnosis and calibration model optimization for near infrared spectroscopy analysis[J].Spectroscopy and Spectral Analysis, 2004, 24(10):1205-1209.
[20] LIU M D, FU R, LIU J C, et al.Quantitative analysis of amorphous form in indomethacin by near infrared spectroscopy combined with partial least squares regression analysis[J].Molecules, 2024, 29(22):5290.
[21] 迟雅丽, 赵亚红, 杨丽, 等.近红外技术用于白砂糖色值检测模型的建立[J].现代食品, 2020, 26(10):195-199.
CHI Y L, ZHAO Y H, YANG L, et al.Near-infrared technology for the establishment of detection model for color of white granulated sugar[J].Modern Food, 2020, 26(10):195-199.
[22] THOMAS E V, HAALAND D M.Comparison of multivariate calibration methods for quantitative spectral analysis[J].Analytical Chemistry, 1990, 62(10):1091-1099.
[23] ZHANG Y, CONG Q, XIE Y F, et al.Quantitative analysis of routine chemical constituents in tobacco by near-infrared spectroscopy and support vector machine[J].Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2008, 71(4):1408-1413.
[24] SUN J Q, YANG X L, ZHOU H X, et al.Rapid quantitative analysis of natural indigo dye content using near-infrared spectroscopy[J].Textile Research Journal, 2024, 94(17-18):1887-1896.
[25] LUO R M, TAN S M, ZHOU Y P, et al.Quantitative analysis of tea using ytterbium-based internal standard near-infrared spectroscopy coupled with boosting least-squares support vector regression[J].Journal of Chemometrics, 2013, 27(7-8):198-206.
Options
文章导航

/

〈 〉

技术支持：北京玛格泰克科技发展有限公司