论文部分内容阅读
针对市场上沙棘籽油质量参差不齐的情况,结合近红外光谱技术研究沙棘籽油快速鉴伪的方法。采用234份沙棘籽油、其他植物油、掺假沙棘籽油的近红外透反射光谱,结合簇类独立软模式法(SIMCA)、偏最小二乘判别法(PLS-DA)、支持向量机法(SVM)3种化学计量学方法,在4 000~6 000 cm-1波段范围内分别建立这3类油的判别模型,并用117份独立样品对模型进行验证。结果表明:3种建模方法均得到了满意的结果,其中SVM在训练和验证过程中均得到100%的正确率,判别效果最好;近红外光谱技术应用于识别纯沙棘籽油和区分沙棘籽油掺假类别具有实用性,近红外光谱技术应用于沙棘籽油鉴伪是可行的。
In view of the uneven quality of sea buckthorn seed oil on the market, combined with near infrared spectroscopy to study the rapid detection of seabuckthorn seed oil method. Near infrared transmission spectra of 234 seabuckthorn seed oil, other vegetable oils and adulterated seabuckthorn seed oil were determined by SIMCA, PLS-DA and support vector machine method, respectively SVM) three kinds of stoichiometry, respectively, in the range of 4 000 ~ 6 000 cm-1 to establish a discriminant model of these three types of oil, and with 117 independent samples to validate the model. The results show that all the three modeling methods have achieved satisfactory results, of which the SVM obtained 100% correctness during training and verification, and the discriminant effect was the best. Near-infrared spectroscopy was used to identify pure seabuckthorn seed oil and seabuckthorn It is practicable to use adulterated seed oil, and it is feasible to use near-infrared spectroscopy in seabuckthorn seed oil testing.