花生油过氧化值和酸值近红外分析模型构建
2018-08-14张欣唐月异王秀贞吴琪孙全喜王志伟刘婷王传堂
张欣 唐月异 王秀贞 吴琪 孙全喜 王志伟 刘婷 王传堂
摘要:为快速测定花生油过氧化值和酸值,保证食用油质量安全,构建了花生油过氧化值和酸值的近红外定量分析模型。采集32份花生油近红外光谱,并按标准方法进行过氧化值和酸值测定。经优化,过氧化值的最佳光谱预处理方法为“一阶导数+矢量归一化”,过氧化值谱区范围为6 094.3~7 506.0 cm-1,维数为6,模型的决定系数(R2)为91.93,交叉验证根均方差(RMSECV)为1.23;酸值最佳光谱预处理方法为“矢量归一化”,酸值谱区范围为6 094.3~7 506.0 cm-1,维数为7,模型R2为93.88,RMSECV为0.074。综上所述,所建模型可以快速准确地预测花生油的过氧化值和酸值。
关键词:花生油;过氧化值;酸值;近红外定量分析
中图分类号:S565.201文献标识号:A文章编号:1001-4942(2018)06-0167-04
Abstract For rapid determination of peroxide value and acid value of peanut oil to ensure quality safety of edible oil, the near infrared spectroscopy models were developed. Near infrared spectrum of a total of 32 peanut oil samples were collected, and the peroxide value and acid value of each sample were determined using standard methods. For peroxide value, the optimized spectral preprocessing method was the first derivative plus vector normalization, and the spectral range, rank, determination coefficient (R2) and root-mean-squares error of cross-validation (RMSECV) were 6 094.3~7 506.0 cm-1, 6, 91.93, and 1.23, respectively. For acid value, the best spectral preprocessing method was the vector normalization, and the spectral range, rank, R2 and RMSECV were 6 094.3~7 506.0 cm-1, 7, 93.88 and 0.074. In conclusion, the models could be used to predict the peroxide value and acid value of peanut oil fastly and accurately.
Keywords Peanut oil; Peroxide value; Acid value; Near infrared spectroscopy (NIRS)
作为人类三大营养素之一,食用油脂是人类膳食的重要组成部分,同时也是食品工业的基础原料。据国家粮油信息中心统计,2016年我国食用油人均消费量为24.8 kg。我国居民主要消费食用油有大豆油、菜籽油、花生油、葵花籽油、玉米油、棕榈油、棉籽油等,其中大豆油、菜籽油、花生油3类占消费量的70%以上,花生油年产量约200万吨 [1]。随着生活水平的提高,人们越来越注重食用油的口感、营养、健康和方便。花生油中不饱和脂肪酸含量约为80%,对人体健康十分有益,而且口感好,深受消费者喜爱。近年来花生油消费量大幅增加,但其质量安全不容忽视。
与其他植物油一样,花生油在加工、流通和保存期间被氧化会导致风味、颜色和质地的劣变[2],产生的氧化脂质会危害人体健康[3]。……
