基于近红外透射光谱的食用油氧化定性分析

为建立食用油氧化的快速鉴别方法，以食用油为材料，根据国家标准对酸价和过氧化值的限量规定将不同氧化程度的样品分为未氧化油和已氧化油两类，通过近红外光谱透射技术结合距离判别分析，建立未氧化油和已氧化油的定性判别模型。结果表明：在波数范围为5450～4650cm-1，采用多元散射校正、一阶求导和Norris导数平滑光谱处理，校正集模型识别率为99%，验证集模型识别率为97.8%，模型预测效果良好，利用近红外透射光谱食用油氧化定性分析可行。

Qualitative Analysis of Edible Oil Oxidation by Near Infrared Transmission Spectroscopy

To establish a quick method for identifying of edible oil oxidation,the qualitative discrimination between oxidized and unoxidized oils as categorized according to the limits of acid value and peroxide value stipulated by the Chinese national standards was modeled based on near infrared transmission spectroscopy(NITS) and distance-discrimination analysis.The results indicated that recognition rates of the calibration and external validation models reached up to 99% and 97.8% in the wave number range from 5450 cm-1 to 4650 cm-1 after multiple scattering correction,first derivative and Norris derivative smoothing.Therefore,good prediction results were achieved.In conclusion,it is feasible to apply NITS to qualitatively analyze edible oil oxidation