论文部分内容阅读
应用主成分分析法(PCA)对多种不同品质油脂样品的LF-NMR弛豫特性数据(T21、T22、T23、S21、S22、S23)进行了分析。研究表明,应用PCA可明显区分正常大豆油和棉籽油,但棕榈油和猪油样品分布范围有一定的重合;无论是轻度煎炸还是深度煎炸,不同种类的煎炸油脂在PCA得分图中能够明显分区;随煎炸时间的延长,油样在PCA得分图上的分布逐渐向左上方移动,四种油脂能够明显分为两类:棕榈油和猪油,大豆油和棉籽油;食用猪油与掺伪猪油样品在PCA得分图上能够明显区分,掺伪比例愈高,二者的区分效果愈好,试验验证正确。说明基于油脂的LF-NMR弛豫特性,结合主成分分析可实现对食用油脂种类、煎炸油程度及掺伪猪油的品质区分。
Principal component analysis (PCA) was used to analyze the LF-NMR relaxation characteristics (T21, T22, T23, S21, S22 and S23) of different oil samples. The results show that the application of PCA can clearly distinguish between normal soybean oil and cottonseed oil, but the distribution range of palm oil and lard has a certain degree of overlap. Whether it is mild or deep frying, the different types of frying fat in the PCA score chart With the extension of frying time, the distribution of oil samples on the PCA score map gradually moves to the upper left and the four types of grease can be obviously divided into two categories: palm oil, lard, soybean oil and cottonseed oil; Lard and adulterated lard samples can be clearly distinguished on the PCA score chart. The higher the adulteration rate, the better the distinction between the two and the correct test. This shows that based on the LF-NMR relaxation characteristics of oils and fats, combined with principal component analysis can distinguish between the types of edible oils and fats, the degree of frying oil and the quality of adulterated lard.