为实现苦荞产地溯源以及确定苦荞成分对苦荞产地溯源的影响程度,对朔州、内蒙古、云贵高原、四川大凉山、陕西5个产地的72个苦荞样本的近红外光谱数据进行了主成分分析(principal component analysis,PCA)和灰色关联分析。结果表明,PCA可以很好地实现不同产地苦荞的聚类,得到的特征波长分别为1 370、1 680、870和971 nm;将上述特征波长与苦荞的6种成分进行灰色关联分析,其灰色关联度由大到小排列为:碳水化合物>蛋白质>脂肪>钠>硒>黄酮;依据关联度大小,从官能团层面确定了碳水化合物和蛋白质是对苦荞产地溯源影响最大的两个成分。表明PCA和灰色关联分析结合近红外光谱技术可以实现苦荞产地溯源研究,为苦荞地理标志产品鉴别提供了一种快速、高效、低成本的方法。
In order to trace the origin of tartary buckwheat and analyze the influence of tartary buckwheat components on its origin traceability, principal component analysis (PCA) and grey relational analysis were carried out on the near infrared spectroscopy data of 72 samples from Shuozhou, Inner Mongolia, Yunnan-Guizhou Plateau, Daliangshan of Sichuan province and Shaanxi province. The results showed that PCA could cluster tartary buckwheat from different areas, and the characteristic wavelengths were 1 370, 1 680, 870 and 971 nm, respectively. Furthermore, grey relational analysis was conducted with the grey relational degree ranking from large to small as follows: carbohydrate > protein > fat > sodium > selenium > flavonoids. As a result, carbohydrate and protein were found to be the two components with greatest influence on the origin of tartary buckwheat at the functional group level. PCA and grey relational analysis combined with near infrared spectroscopy can be used to trace the origin of tartary buckwheat, providing a fast, efficient and low-cost method for the identification of tartary buckwheat geographical indications.
[1] 郑峰,孙文文,张琦,等.苦荞籽粒的化学成分研究[J].西北农林科技大学学报(自然科学版),2011,39(10):199-203.
[2] 田秀英,李会合.苦荞籽粒蛋白质营养评价及其对硒的响应[J].食品科学,2010,31(7):105-108.
[3] 谢倩,黄文耀,李建洪,等.硒蛋白中硒代氨基酸形态检测技术进展[J].公共卫生与预防医学,2015,26(3):63-66.
[4] 韩雍. 荞麦芦丁和蛋白质的近红外光谱分析[D].杨凌:西北农林科技大学,2008.
[5] WATANABE A, FURUKAWA H, MIYAMOTO S, et al.Non-destructive chemical analysis of water and chlorine content in cement paste using near-infrared spectroscopy[J]. Construction and Building Materials,2019,196:95-104.
[6] CHEN Hui, TAN Chao, LIN Zan.Non-destructive identification of native egg by near-infrared spectroscopy and data driven-based class-modeling[J]. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy,2018,206:484-490.
[7] 马续莹,张卓勇.近红外光谱法在燕麦和苦荞膳食纤维含量分析中的应用[J].首都师范大学学报(自然科学版),2011,32(5):30-33.
[8] 席志勇. 基于近红外光谱技术荞麦无损检测方法研究[D].昆明:昆明理工大学,2013.
[9] 张晶,郭军,张美莉. 荞麦掺假近红外快速检测模型的建立[J].食品工业,2018,39(10):195-199.
[10] 梁润梅. 硒蛋白及其生物学功能[J].生物学教学,2004,36(9): 5-7.
[11] 杰尔·沃克曼,洛伊斯·文依.近红外光谱解析实用指南[M].北京:化学工业出版社,2009:274.
[12] ACETO M, ROBOTTI E, ODDONE M, et al.A traceability study on the Moscato wine chain[J]. Food Chemistry, 2013, 138(2-3):1 914-1 922.
[13] KELLY S D, HEATON K, HOOGEWERFF J.Tracing the geographical origin of food: the application of multielement and multi-isotope analysis[J]. Trends in Food Science and Technology, 2005,16(12): 555-567.
[14] 王世芳. 贮藏期番茄果实质地的光谱特性分析[D].太原:山西农业大学,2016.
[15] 卢艳丽,白由路,杨俐苹,等.基于主成分回归分析的土壤有机质高光谱预测与模型验证[J].植物营养与肥料学报,2008,14(6):1 076-1 082.
[16] 臧卓,林辉,杨敏华.ICA与PCA在高光谱数据降维分类中的对比研究[J].中南林业科技大学学报,2011,31(11):18-22.
[17] 张玥,王朝辉,张亚婷,等. 基于主成分分析和判别分析的大米产地溯源[J].中国粮油学报,2016,31(4):1-5.
[18] 雷建刚,刘敦华.近红外光谱结合SIMCA对不同产地枸杞溯源模型的优化[J].食品科学,2013,34(20):148-152.
[19] 颜静,唐成,梁亚雄,等.柚子原产地溯源鉴定技术[J].食品科学,2014,35(4):248-252.
[20] 苏学素,张晓焱,焦必宁,等.基于近红外光谱的脐橙产地溯源研究[J].农业工程学报,2012,28(15):240-245.
[21] 马国峻,王水波,裴庆祺,等.基于主成分分析和K-means聚类的平行坐标可视化技术研究[J]. 网络与信息安全学报,2017,3(8):18-27.
[22] 王海峰,张智韬,ARNON KARNIELI,等. 基于灰度关联-岭回归的荒漠土壤有机质含量高光谱估算[J].农业工程学报,2018,34(14):124-131.
[23] 李明亮,李西灿,张爽. 土壤含水量高光谱灰色关联度估测模式[J]. 测绘科学技术学报,2016,33(2):163-168.
[24] 宋雪健,钱丽丽,张东杰,等.近红外光谱技术在食品溯源中的应用进展[J].食品研究与开发,2017,38(12):197-200.
[25] 张兰勇,李陇南,陈辉煌.采用灰色关联分析方法评估火箭弹毁伤效能[J].兵器装备工程学报,2018,39(5):11-15.
[26] 宋海燕,王世芳,谌英敏,等.基于灰色关联度和Vis-NIR的不同贮藏方式下番茄光谱特性分析[J].山西农业大学学报(自然科学版),2019,39(2):75-78.
[27] ZHU Y, WEINDORF D C,CHAKRABORTY S.Characterizing surface soil water with field portable diffuse reflectance spectroscopy[J]. Journal of Hydrology, 2010, 391(1):133-140.
[28] 杨玮,孙红,郑立华,等.冬枣光谱数据的灰色关联分析及叶片氮素含量预测[J].光谱学与光谱分析,2013,33(11):3 083-3 087.
[29] HUANG Yuansheng, SHEN Lei, LIU Hui.Grey relational analysis, principal component analysis and forecasting of carbon emissions based on long short-term memory in China[J]. Journal of Cleaner Production,2019, 209:415-423.
[30] 崔凯. 基于模型合并的无监督降维方法研究[D].西安:西安电子科技大学,2014.