应用傅立叶变换近红外光谱检测婴幼儿配方奶粉中的乳糖含量

利用傅立叶变换近红外光谱（FT-NIR）技术结合偏最小二乘法（PLS）对婴幼儿配方奶粉中的乳糖进行快速检测分析。搜集 94个不同产地、不同品牌的婴幼儿配方奶粉中乳糖的实验室数据，并采集婴幼儿配方奶粉的近红外光谱图，选择最优的光谱预处理方法，优化、验证和建立模型，并预测6个未知品牌的婴幼儿配方奶粉样品的乳糖含量。结果表明，不同阶段的婴幼儿配方奶粉有非常相似 的近红外特征图谱，图谱处理后的方差为95.423 5%，该模型测定的乳糖值与高效液相色谱（HPLC）法测定的乳糖值之间的平均相对误 差均≤0.67%，相对标准偏差（RSD）均≤0.88%，均符合误差范围。该方法可以无损、快速、高效地测定婴幼儿配方奶粉中的乳糖含量。

Determination of lactose content in infant formula milk powder based on Fourier transform near-infrared spectroscopy

Th e lactose content in infant formula milk powder was rapidly detected and analyzed by Fourier transform near-infrared spectroscopy (FT-NIR) combined with partial least squares (PLS). 94 samples of infant formula milk powder from different originals and brands were collected, and NIR spectra of all infant formula milk powder were collected. The optimal pre-treatment method of spectra was chosen, and then the model was optimized, verified, and established, and the lactate content of 6 samples of unknown infant formula milk powder were forecasted. The results showed that different infant formula milk powder had the same NIR spectrum, the variance of the spectra after treatment was 95.423 5%, the average relative errors of lactose values determined by the model and HPLC method were less than 0.67%, the relative standard deviation (RSD) was less than 0.88%, and all of them were conformed to the error range. The results showed that method could determine lactate content in infant formula milk powder nondestructively, rapidly and efficiently.