【摘要】 目的 建立基于近红外光谱快速测定食用植物油中酸价的分析方法。方法 采用冷溶剂指示剂滴定法检测371个食用植物油样品的酸价,并采集样品的近红外光谱。经过标准正态变换结合一阶导数对近红外光谱进行数据预处理,选用竞争性自适应重加权采样算法选取重要变量,建立食用植物油酸价的偏最小二乘回归模型。结果 蒙特卡洛交互验证结果显示,食用植物油酸价预测模型的验证集决定系数Q2为0.9983,交互检验的均方根误差(root mean square error of cross validation, RMSECV)为0.0461,模型预测的独立测试集的酸价与实测值相关系数为0.9834,预测效果良好。结论 本研究建立的食用植物油酸价近红外光谱快速检测方法能够满足检测要求,为评价或跟踪食用油品质提供快速无损的技术思路。更多还原

【Abstract】 Objective To establish a method for the rapid determination of acid value of edible vegetable oil based on near-infrared spectroscopy. Methods The acid values of 371 edible vegetable oil samples were detected by cold solvent indicator titration. Meanwhile, near-infrared spectra of these samples were collected. After preprocessing including standard normal variable transformation and first-order derivation, competitive adaptive re-weighted sampling was used for wavelength selection, and then a partial least square regression model was established to predict the acid value of edible vegetable oil. Results Monte Carlo cross validation results showed that determination coefficient(Q2) of this model was 0.9983 for the validation set, and the root mean square error of cross validation(RMSECV) was 0.0461. Moreover, correlation coefficient of the predicted acid values and the measured ones in the independent validation set was 0.9834, indicating the prediction effect was good. Conclusion The fast detection method of acid value of edible vegetable oil by near-infrared spectroscopy established in this study can meet the detection requirements, and provide a rapid and nondestructive technical idea for evaluating or tracking the quality of edible oil.更多还原