9种植物油脂Fisher判别函数的表征研究

Study on the Characterization of 9 Kinds of Vegetable Oils Fisher Discriminant Function

本研究为了建立不同植物油脂的识别方法,针对9种植物油脂采用气相色谱法分析其脂肪酸组成,用面积归一化法定量各脂肪酸甲酯的含量。以豆蔻酸、棕榈酸、硬脂酸、油酸、亚油酸、亚麻酸、花生酸、花生烯酸和芥酸等9种脂肪酸为变量,应用Fisher判别法,建立8个判别函数,以第一和第二个判别函数得分绘制判别得分二维散点图,9种油脂聚成9类,并且原始判别和交叉验证的准确率均为100%。另外,比较不同判别函数下各类植物油脂的重心坐标可知,判别函数Y1下的重心坐标两两具有明显差别,故选取判别函数Y1作为外部验证判别函数。通过外部数据的验证显示,使用判别函数Y1能够准确判定未知样的归属,方法选择9种主要脂肪酸作为变量,具有一定的可行性。

For establishing identification methods of different vegetable oils, gas chromatogram is applied to analyze fatty acid methyl esters of 9 kinds of vegetable oils, and the content of which is quantified by peak - area normalization method. Myristic acid, palmitic acid, stearic acid, oleic acid, linoleic acid, linolenic acid, arachidic acid, arachidonic acid, erucic acid are chosen as variables, and Fisher discriminant is used to estabilish 8 discriminant functions. Draw 2D scatterplot chart about discriminant score based on the scores of the first and second discriminant function. 9 kinds of oils accumulate to 9 classes and the precision of original discrimination and cross validation is 100% respectively. Besides, comparing group centroids of vegetable oils from different functions, the ones of the first function is obviously odds from each other, therefore the first function is chosen as discriminant function for external authentica- tion. The fatty acids of some unknown samples are brought into the first function. The result shows that the first function can distinguish validation samples with belongings exactly. The method is certain feasible to discriminate 9 kinds of vegetable oils and fats with 9 kinds of fatty acids as variables for functions.