基于近红外光谱对婴幼儿配方奶粉中非法添加物的快速鉴别方法

利用近红外光谱技术结合Adulterant Screen算法建立婴幼儿配方奶粉中非法添加物快速鉴别方法。采集婴幼儿配方奶粉样品及非法添加物近红外光谱,建立婴幼儿配方奶粉分类模型及掺假物模型,运用Adulterant Screen算法进行计算分析,建立了婴幼儿配方奶粉中非法添加物的快速鉴别方法。方法对单组份掺杂婴幼儿配方奶粉的识别最低检出浓度为:三聚氰胺0.04%、环丙氨嗪0.05%,尿素、双缩脲及硫脲0.1%,双氰胺0.2%。对于二组分和三组分掺假奶粉在两倍到十倍检出限浓度的准确识别率分别达到了83.3%和50%,对无掺假样品的识别率100%,同时,该鉴别模型可以通过一个模型即可实现快速鉴别多个非法添加物。本研究建立的模型能为婴幼儿配方奶粉提供了一种操作简单、快速、可靠、样品无损的掺假鉴别方法,能有效的运用于掺假鉴别的日常检测工作,为婴幼儿配方奶粉的检测和监管提供强有力的技术支持。

Rapid Method for the Identification of Illegal Additives in Infant Formula Based on Near Infrared Spectroscopy

The rapid identification method was established by using near infrared spectroscopy combined with Adulterant Screen algorithm to identify the illegal additive in infant formula.Infant formula and illegal additives near-infrared spectrum were collected to establish the formula classification model and adulteration model.Using the Adulterant Screen algorithm to calculation and analysis,establishes the infant formula in the rapid identification of illegal additives.The minimum detection concentration of methods of single component adulterated infant formula was as follows:Melamine 0.04%,cyromazine 0.05%,urea,biuret and thiourea 0.1%,dicyandiamide 0.2%.The accurate identification rate of two-component and three-component adulterated infant formula at the detection limit concentration of two to ten times were 83.3%and 50%.The identification rate of unknown samples without adulteration was 100%.At the same time,the identification model could quickly identify multiple illegal additives through one model.The model established in this study could provide a simple,fast,reliable,nondestructive adulteration identification method for infant formula.It could be effectively used in the daily detection of adulteration identification of infant formula and provides strong technical support for the detection and supervision of infant formula.