Chemometric Methods Applied Research In The Olive Oil Quality Analysis And Identification Of Wheat

Olive oil is made from ripe olives without by chemical treatments, which change its structure and consequently its physicochemical and functional properties. It is digested easily. It's not only used as a nice kind of edible oil, but also used for beauty and the prevention of heart and brain disease. Now olive oil is salable in international market, so it is sometimes adulterated with similar cheap edible oil by some greedy business people and supplier. They are quite alike. It is difficult to distinguish the pure olive oil and the mixtures by eyes. There are several different methods used in the detection of adulteration, including UV, HPLC, NMR and florescence spectrometry. These techniques are expensive and time-consuming. So it needs to develop a quick, easy, efficient method.Oat and buckwheat have a high content of biologically valuable proteins, high portion of lipids compared to other cereals, vitamins, mineral substances and dietary fiber. It has been shown that oat and buckwheat can reduce serum lipids and blood sugar in humans, make people lose weight and so on. They are more and more popular in sale market for nutrition and functional value. Oat and buckwheat seeds can be easily distinguish by their shapes, but it is difficult to tell when they are powder. So it is feasible to develop a simple, quick method. The fast and non-destructive near-infrared spectra (NIRS) are suitable for the classification of olive oils and grist. But NIRS is an indirect method of measurement, and it needs analysis by many chemometrics methods.In this research, chemometrics methods are applied in the classification of olive oils and grist. The results as follows:1. Qualitative and quantitative analysis of olive oils by near-infrared spectraNIRS of olive oils are studied by Discriminant analysis. The best wave bands for analysis are selected. The near-infrared spectra and first-derivative spectra of olive oils are used for analysis, respectively. Furthermore, NIRS of olive oils are studied by support vector machine (SVM). The near-infrared spectra of olive oils are processed by smooth, compress and normalization. The result is satisfying. The quantities of adulteration in pure olive oils are studied by partial least square (PLS). The prediction results show that the components of camellia oil and poppy seed oil are more similar with olive oil, compared with colza oil, corn oil, peanut oil, and sunflower oil. The conclusion is that the combination of NIRS and chemometrics is a fast and convenient method for qualitative and quantitative analysis of olive oils.2. Qualitative analysis of oat and buckwheat by Discriminant analysis and SVMNIRS of oat and buckwheat is classified by Discriminant analysis. NIRS of oat and buckwheat samples are collected under two conditions, impaction and without impaction. The near-infrared spectra and first-derivative spectra of samples are studied, respectively. Four different kernel functions of SVM are applied to develop classification models. The parameters affecting four different kernel functions prediction are discussed. The prediction results prove that NIRS is a reliable method for the identification oat and buckwheat. Moreover, if more samples are obtains, the prediction values would be better.