近红外光谱分析技术在植物蛋白饮料定量分析中的应用

发布时间：2018-10-05 17:03

【摘要】：利用近红外光谱分析技术对植物蛋白饮料中脂肪和可溶性固形物含量进行定量分析。采用向后间隔偏最小二乘法(Bi PLS)、组合间隔偏最小二乘法(Si PLS)、遗传偏最小二乘法(GA-PLS)、竞争性自适应重加权法(CARS)优选波段,并结合偏最小二乘法(PLS)建立植物蛋白饮料中脂肪和可溶性固形物的定量分析模型。结果表明,4种方法对模型均有优化效果,可提高模型的稳定性和精准性,其中GA-Bi PLS、GA-Si PLS优化效果最为明显,脂肪、可溶性固形物的决定系数R2分别达到了0.984、0.97和0.988、0.990,预测标准均方差(RMSEP)分别为0.026、0.030和0.170、0.155,相对分析误差(RPD)分别为8.077、7.000和9.112、10.000。表明近红外光谱技术作为一种快速、便捷的检测手段,适用于植物蛋白饮料品质的快速检测分析。
[Abstract]:The content of fat and soluble solids in vegetable protein beverage was quantitatively analyzed by near infrared spectroscopy (NIR). The backward interval partial least squares (Bi PLS),) combined interval partial least squares (Si PLS),) genetic partial least squares (GA-PLS) method and competitive adaptive reweighting method (CARS) were used to optimize the band selection. The quantitative analysis model of fat and soluble solids in vegetable protein beverage was established by partial least square method (PLS). The results show that all of the four methods can improve the stability and accuracy of the model, and the optimization effect of GA-Bi PLS,GA-Si PLS is the most obvious. The coefficient of determination (R2) of soluble solids was 0.9840.97 and 0.9880.900.The standard mean deviation (RMSEP) of the prediction was 0.026 ~ 0.030 and 0.170 ~ 0.155, respectively. The relative analysis error (RPD) was 8.077 ~ 7.000 and 9.112 ~ 10.000, respectively. Near-infrared spectroscopy is a rapid and convenient method to detect and analyze the quality of vegetable protein beverage.
【作者单位】：中国食品发酵工业研究院;东北农业大学工程学院管理科学工程系;河北养元智汇饮品股份有限公司;红牛维他命饮料有限公司;
【基金】：国家自然科学基金项目(31671937)
【分类号】：O657.33;TS275.4

【相似文献】