真空油炸马铃薯脆片预处理工艺优化

Optimization of the Vacuum Fried Potato Chips Pretreatment Technique

为优化马铃薯脆片的预处理工艺,采用单因素和响应面试验,以破碎力、含油量、L*值、感官评分、综合评分和电子鼻检测挥发性成分为指标,对实验数据进行主成分分析。结果表明:不同的预处理方式对马铃薯脆片的破碎力、感官评价和综合评分有显著影响(p<0.05)。主成分分析共提取5个主成分,前3个主成分总贡献率大于85%,说明提取3个主成分能够全面反映马铃薯脆片的品质信息,决定第1主成分的指标主要是感官评价和综合评分,决定第2主成分的是破碎力和含油量;决定第3主成分的是L*值和含油量。以主成分分析得到的规范化综合得分为响应值建立的二次多项式回归模型回归效果极显著(p<0.01,R^2=0.9604)。确定最佳工艺参数为漂烫温度91℃、漂烫时间4 min、切片厚度4 mm和冷冻时间3 h,在此条件得到规范化综合评分0.9572,与预测值(0.9453)相接近,表明以主成分分析得到的规范化综合评分为响应值建立的回归模型具有良好的预测能力。

In order to optimize pretreatment technology of the vacuum fried potato chips, principal component analysis was performed on the experimental data obtained by single factor and response surface tests, including crushing force, oil content, L* value, sensory score, comprehensive score and electronic nose-detected volatile components. The results showed that different pretreatment methods had significant effects on crushing force, sensory evaluation and comprehensive score for potato chip(p<0.05). Principal component analysis showed that 5 principal components were extracted, and the total contribution of the first three principal components was more than 85%. It showed that the three principal components could reflect the quality information of potato chips comprehensively. The first principal component is mainly the sensory evaluation and the comprehensive score, and the second principal component is the crushing force and the oil content;and the third principal component is the L* value and the oil content.The regression effect of the quadratic polynomial regression model established by the comprehensive score in principal component analysis was extremely significant(p<0.01, R^2=0.9604). The optimum process parameters were determined as the blanching temperature of 91 ℃, the blanching time of 4 min, the slice thickness of 4 mm and the freezing time of 3 h. Under these conditions, the normalized comprehensive score was 0.9572, which was close to the predicted value(0.9453), indicating that the normalized comprehensive score obtained by principal component analysis has a good predictive ability for the regression model.