基于多元统计分析方法建立北方馒头品质评价体系

Development of the Quality Evaluation System of Chinese Northern Steamed Bread by Multivariate Statistical Analysis

以17种市售北方馒头为研究对象,基于多元统计分析方法研究其物性指标和感官评价指标间的相互关系,进而建立了适用于北方馒头的品质评价体系。通过对原料的物性指标进行系统聚类分析发现,当相似系数为0.90时,硬度、咀嚼度、胶着性可归为一类,故选用硬度代替咀嚼度与胶着性。通过偏最小二乘法分析与相关性分析,研究了北方馒头物性指标和感官评价指标间的相互关系,结果表明硬度、弹性、比容、径高比及色差L值可作为北方馒头品质评价体系中的关键物性指标,并利用主成分分析方法得到以关键物性指标为基础的综合得分公式Z=0.115X_1+0.452X_2+0.391X_3+0.421X_4+0.017X_5(X_1:硬度; X_2:弹性; X_3:比容; X_4:径高比; X_5:L;),用于北方馒头的物性品质的评价。此外,选取了10种市售北方馒头进行验证,可知该品质评价体系与感官评价体系的结果基本一致,进而说明基于多元统计学方法的北方馒头品质评价体系具有良好的可行性。

This study was established the quality evaluation system of Chinese northern steamed bread through 17 kinds of commercially Chinese northern steamed bread as research object to investigate the relationship between physical properties and sensory characteristics by multivariate statistical analysis.The cluster analysis of physical properties of raw materials showed that when the similarity coefficient was 0.90,hardness,chewiness and gumminess were classified into one class,and hardness was selected to replace the chewiness and gumminess.The relationship between physical properties and sensory characteristics of Chinese northern steamed bread by partial least squares analysis and correlation analysis,the results showed that hardness, elasticity,specific volume,diameter ratio and color difference L value could be used as the key physical items in the quality evaluation system.Based on the comprehensive quality evaluation score formula Z =0.115X 1 +0.452X 2 +0.391X 3 +0.421X 4 + 0.017X 5 (X 1 :hardness,X 2 :elasticity,X 3 :specific volume,X 4 :diameter ratio,X 5 :L value),the quality evaluation system was established by the principal component analysis of the five physical properties,which was suitable for Chinese northern steamed bread.In addition,another ten kinds of commercially Chinese northern steamed bread were selected for verification.The results showed that the quality evaluation system was in good agreement with the sensory evaluation system,which showed that the steamed bread quality evaluation system based on multivariate statistical method was feasible.