基于电子鼻分析鱼油储藏过程中酸败程度的研究

Analysis of Degree of Rancidity of Fish Oil During Storage by Using Electronic Nose

鱼油的酸败往往伴随着气味的变化,利用电子鼻对储藏过程中金枪鱼油的挥发性气味进行研究,每5 d测定1次,运用主成分分析(PCA)、线性判别分析(LDA)区分不同储藏时间的鱼油样品,并通过偏最小二乘法(PLS)建立酸价、过氧化值的预测模型。结果表明:随着储藏时间的延长,金枪鱼油整体气味呈增加趋势,这种气味变化趋势与酸价、过氧化值的变化趋势基本一致;不同储藏时间的金枪鱼油样品可以区分开来,LDA分析方法明显优于PCA分析方法;利用PLS偏最小二乘法(PLS)预测金枪鱼油酸价和过氧化值,得到线性模拟方程,验证试验表明,酸价、过氧化值预测值与实测值的平均相对偏差分别为10.60%、12.04%,准确性较好,可作为酸价、过氧化值预测的辅助手段。

The rancidity of fish oil was often accompanied by of odor of tuna oil during nalysis （PCA）, linear di storage using electronic nose, which was scriminant analysis （LDA） to distinguish changes in odor, so this article studied the change measured every 5 d, using principal component athe sample with different storage time and establish forecast model of acid value, peroxide value by partial least squares （PLS）. The results showed that odor of tuna oil was in an increasing trend with the extension of storage time, basically the same as trends of acid value and peroxide value. Tuna oil samples with different storage time could be distinguished and LDA method was better than PCA. PLS was used to forecast acid value and peroxide value of tuna oil, getting linear simulation equationhes. The validation test showed that the average relative deviation between the expected value and actual value of acid value and peroxide value was 10.60% and 12.04%, respectively, which could be used as an aid method of predicting acid value, peroxide value of fish oil during storage.