枇杷蜂蜜中饴糖的近红外光谱检测  

Detection of maltose adulteration in loquat honey by near infrared spectroscopy

作　　者：魏小梅[1] 邵玉芳[1] 应寻源[1] 

机构地区：[1]浙江大学生物系统工程与食品科学学院,浙江杭州310058

出　　处：《食品工业科技》2012年第10期85-87,100,共4页Science and Technology of Food Industry

基　　金：国家公益性行业(农业)专项(200903009)

摘　　要：实验通过对纯枇杷蜂蜜及主动掺入1%、2%……30%饴糖的假枇杷蜂蜜进行近红外光谱扫描,采用TQAnalysisv6对数据进行预处理,建立饴糖含量的定性及偏最小二乘法和主成分回归法定量分析模型,并将模型应用于蜂蜜样品的分析预测。结果显示,采用原始光谱或一阶微分处理建立的判别分析模型均能够较好地区分掺饴糖蜂蜜与纯蜂蜜。根据PLS算法、PCR算法建立的定量模型相关系数分别为0.99771、0.98654,用于预测的蜂蜜样品实际值与预测值之间的决定系数分别为0.992、0.974。由此可见,用近红外光谱技术鉴别蜂蜜中是否添加饴糖是可行的,在实际操作中可以采用近红外光谱法快速定性判别蜂蜜中是否含有饴糖,也可根据化学计量法确定饴糖的含量,为蜂蜜打假提供依据。Near infrared spectroscopy（NIR）was used to detect pure loquat honey（30 samples）and loquat honey adulterated with maltose（120 samples at 30 different rates as 1%,2%…30%）.Spectral datas were first pretreated by TQ Analyst v6,and a qualitative model was established to discriminate the two sorts of honey,then partial least squares（PLS）model and principal component regression（PCR）model were established to determine maltose content in honey.And finally models were applied to predict maltose content in honey samples.Results indicated that both of the discriminant analysis models（one was raw spectrum and one was pretreated by first derivative）could discriminate the two sorts of honey clearly.The correlation coefficient of PLS and PCR models were 0.99771 and 0.98654,respectively.R2 between the experimental values and simulated values of the honey samples for prediction of PLS and PCR models were 0.992 and 0.974,respectively.It could be concluded that NIR could not only detect maltose in honey,but also determine maltose content in honey according to chemometrics methods,which would provide a basis for cracking down on honey adulteration.

关 键 词：蜂蜜 饴糖 近红外光谱 偏最小二乘法 主成分回归法 

分 类 号：TS207.3[轻工技术与工程—食品科学]