近红外光谱分析技术在植物蛋白饮料定量分析中的应用

摘要

The contents of fat and soluble solids in vegetable protein drink were quantitatively analyzed by near infrared spectroscopy.The optimum waveband was selected from the full spectrum using backward interval partial least square (BiPLS),synergy interval partial least square (SiPLS),genetic algorithm partial least square (GA-PLS) and competitive adaptive reweighted sampling (CARS),and the quantitative analysis model of fat and soluble solids in vegetable protein drink was established by partial least square (PLS).The results showed that the four methods had optimization effect on the model,and could improve the stability and precision of the model,in which the optimization effect of GA-BiPLS and GA-SiPLS were the most obvious.The determination coefficient R2 of fat and soluble solids were 0.984,0.97 and 0.988,0.990,the root mean square error of prediction (RMSEP) were 0.026,0.030 and 0.170,0.155,the relative percent deviation (RPD) were 8.077,7.000 and 9.112,10.000,respectively.The research indicated that near infrared spectroscopy was a fast and convenient detection method for rapid detection and analysis of vegetable protein drink quality.%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(BiPLS)、组合间隔偏最小二乘法(SiPLS)、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-BiPLS、GA-SiPLS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.