摘要
为检测油茶籽油的掺伪程度,对5种食用油及两组油茶籽油掺杂混合油进行了介电谱测量。运用主成分分析法(PCA)开展了5种食用油的区分分析,结果表明PCA法不仅能够显著区分单不饱和脂肪酸(MUFA)型和多不饱和脂肪酸(PUFA)型食用油,对于同一类型的不同油品也具有良好的区分度。运用基于交叉验证的偏最小二乘法(PLS),对两种油茶籽油掺杂混合油的介电谱数据建立定量分析模型,经外部验证集验证,混合油的预测均方根误差(RMSEP)小于2.10%,决定系数(R2)大于0.998 9。最后进行了油茶籽油介电谱的温度特性测试,给出了温度补偿的相应算法。试验结果表明,介电谱法为食用油的掺伪鉴别和纯度检测提供了一种快捷、准确的方法。
To detect adukeration of virgin camellia oil, the dielectric spectroscopy of five kinds of edible oils and two kinds of hybrid camellia oils were detected. Principal component analysis （PCA）was applied to distinct different oils, and the results showed that it can be clearly distinguished between the group oil of monounsaturated fatty acids （MUFA） and that of polyunsaturated fatty acid（ PUFA）, and the different oils of the same group also can be identi- fied. Quantitative models of two kinds hybrid camellia oils were established using cross - validation partial least squares （PLS） , the models were verified by external validation set and presented a good prediction capacity with Root Mean Square Prediction Error（RMSEP） less then 2.10% and R - square better than 0.998 9. To provide the temper- ature compensation basis, the relationship between dielectric spectroscopy and temperature was studied, and then give the corresponding algorithm. All showed that the dielectric spectroscopy was a rapid, accurate identification and purity detection for edible oil analysis.
出处
《中国粮油学报》
EI
CAS
CSCD
北大核心
2012年第12期111-115,121,共6页
Journal of the Chinese Cereals and Oils Association
关键词
介电谱法
油茶籽油
主成分分析法
偏最小二乘法
温度特性
dielectric spectroscopy method, camellia oil, principal component analysis, partial least squares,temperature characteristic
