【摘要】 以荆州鱼糕为研究对象，采用感官评定、挥发性盐基氮(TVB-N)检测和电子鼻技术分析不同贮藏条件下荆州鱼糕的新鲜度变化，并基于电子鼻检测数据，结合主成分分析(PCA)、层次聚类分析(HCA)、偏最小二乘判别分析(PLS-DA)和逐步多元线性回归分析(Stepwise-MLR)建立鱼糕新鲜度的判别和预测模型。结果表明：在4℃和室温条件下，电子鼻响应信号均能很好地区分鱼糕样品的新鲜度；鱼糕贮藏过程中产生的氮氧化物、硫化物、甲烷等是新鲜度下降的重要指标；基于电子鼻检测数据建立的多元线性回归预测模型的R~2均大于0.932 5,预测集样品的预测均方根误差均小于1.22,即电子鼻技术结合多元统计分析可作为一种无损、简便和快速检测鱼糕新鲜度的方法。更多还原

【Abstract】 At different storage conditions, the freshness of fish cake after storage was analyzed by sensory evaluation, volatile basic nitrogen(TVB-N) detection and electronic nose. Combining with principal component analysis(PCA), hierarchical cluster analysis(HCA), partial least square discriminant analysis(PLS-DA) and stepwise multiple linear regression analysis(Stepwise-MLR), freshness discriminant and predictionanalysis based on the electronic nose data were carried out.The research results show that the freshness of the fish cake samples was well distinguished at 4 ℃ and room temperature. Nitrogen oxide, sulfide and methane were the important indicatiors of freshness decrease. The correlation coefficients between measured and predicted TVB-N values based on stepwise-MLR were greater than 0.932 5. And the predicted root mean square errors of the prediction set samples were less than 1.22. It could be concluded that electronic nose combined with multivariate statistical analysis could provide a nondestructive, simple and rapid method for detecting the freshness of fish cake.更多还原