论文部分内容阅读
目的应用傅里叶变换红外光谱(FTIR)结合最小偏二乘法(PLS)建立大豆原油-棕榈油二元掺伪体系的定量分析模型。方法以42个大豆原油、21个精炼油、88个掺伪油的FIIR谱图为模型样本,预处理方法选用标准正态变量(SNV),在此基础上应用主成分分析(PCA)提取特征变量,随机选取60个掺伪油样组成校正集,28个掺伪油样组成验证集,以PLS方法建立大豆原油的掺伪定量模型。结果 PCA可将大豆原油及精炼油分成独立的2类。经PCA分析,大豆原油中掺入棕榈油的掺伪检测限为5%。PLS校正模型的判定系数R2为0.9926,校正误差均方根RMSEC为1.8121。预测模型的R2为0.9823,交叉验证误差均方根RMSECV为2.8189。同时得到的预测结果的偏差在1.3909%~3.1019%之间,差异不显著,说明此模型可行。结论 FTIR-PLS模型能够实现大豆原油的掺伪定量分析,分析速度快,能够满足大豆原油入库要求,是一种可行的大豆原油掺伪分析方法。
OBJECTIVE To establish a quantitative model for the binary adulteration of soybean crude oil and palm oil by using Fourier transform infrared spectroscopy (FTIR) and least square partial least squares (PLS). Methods The FIIR spectra of 42 soybean oil, 21 refined oil and 88 adulterated oil were used as the model samples. The standard normal variable (SNV) was used as the pretreatment method. PCA (principal component analysis) Variables, randomly selected 60 sets of adulterated oil samples to make up the calibration set, and 28 adulterated oil sample sets to verify the set. The adulteration quantitative model of soybean crude oil was established by PLS method. Results PCA separates soybean crude oil and refined oil into two separate categories. Analyzed by PCA, the adulterated detection limit of soybean oil blended with palm oil is 5%. The coefficient of determination R2 of the PLS calibration model is 0.9926, and the RMSEC of the correction error is 1.8121. The R2 of the prediction model is 0.9823, and the root mean square RMSECV of cross validation error is 2.8189. The deviation of the prediction results obtained at the same time ranged from 1.3909% to 3.1019%. The difference was not significant, indicating that this model is feasible. Conclusion The FTIR-PLS model can realize the quantitative analysis of soybean crude oil, which is fast and can meet the requirements of soybean crude oil storage. It is a feasible method for soybean oil adulteration analysis.