基于主成分分析果桑品质评价模型的建立  

Establishment of quality evaluation model of fruit mulberry based on principal component analysis

作　　者：史星雲 徐珊珊 柳丽萍 陶新中 钱文春 SHI Xingyun;XU Shanshan;LIU Liping;TAO Xinzhong;QIAN Wenchun(Huzhou Academy of Agricultural Science,Zhejiang 313000;Shuangin Town Public Service Center;Key Laboratory of Huzhou Agricultural Germplasm Resources Innovation and Application;Huzhou Lianshi Zhujiadou Sericulture Professional Cooperative)

机构地区：[1]湖州市农业科学研究院,浙江313000 [2]双林镇公共事业服务中心 [3]湖州市农业种质资源创新与应用重点实验室 [4]湖州练市朱家兜蚕桑专业合作社

出　　处：《中国果树》2024年第3期54-59,共6页China Fruits

基　　金：浙江省产业技术团队项目(2022CSSFD08);财政部和农业农村部:国家现代农业(蚕桑)产业技术体系湖州综合试验站(CARS-18-SYZ05)。

摘　　要：为探明浙江省湖州区域果桑品质特征并构建其品质评价模型,以粤椹大10果实为材料,测定其单果鲜重、含水率等12项品质指标,运用描述性统计、相关性、主成分和回归分析等方法对其进行分析和评价,并建立其品质评价模型。结果表明,不同样品间果实品质存在差异,不同指标离散程度不同,变异系数为2.28%~35.56%。相关性分析表明,12项指标间存在不同程度的正相关或负相关。运用主成分分析提取3个主成分,累计方差贡献率为89.723%,其中第1主成分为保健功能因子,方差贡献率为53.172%;第2主成分为风味因子,方差贡献率为25.205%;第3主成分为重量因子,方差贡献率为11.346%,主成分分析结果可以很好地区分不同样品间的品质差异。最终采用逐步回归分析方法建立了粤椹大10品质评价预测模型,筛选出了花色苷等6个有效指标。综上,粤椹大10品质评价模型为:Y=-9.791+0.356X_(12)+0.653X_(9)+3.027X_(2)+0.051X_(7)+0.468X_(10)+0.011X_(8),其中花色苷、维生素C、单果干重、固酸比、黄酮和总糖等指标是评价其品质的重要指标,运用主成分分析方法可以评价该品种果实品质,可为粤椹大10品质评价体系构建提供理论依据。To explore the quality characteristics of fruit mulberry in Huzhou district of Zhejiang Province,and construct its quality evaluation model.‘Yueshenda 10’were used as materials and determined 12 fruit quality indices.The methods of descriptive statistical analysis,correlation analysis,principal component analysis and regression analysis were conducted,and then the quality evaluation model was established.The results showed that there were differences in fruit quality among different samples,and the degree of dispersion between 12 indicators was different.The coefficient of variation was from 2.28%to 35.56%.Correlation analysis showed that there were different degrees of correlation among the 12 indicators.Three principal components were obtained with a cumulative contribution rate of 89.723%after principal component analysis.The first principal component was health factor with variance contribution approached to 53.172%.The second principal component was flavor factor with variance contribution approached to 25.205%.The third principal component was fruit weight factor with variance contribution approached to 11.346%.The comprehensive scores could well distinguish the quality differences between different samples.Finally,we established the quality evaluation prediction model of‘Yueshenda 10’by the stepwise regression analysis,and screened 6 effective indexes such as anthocyanins,etc.The quality evaluation model of‘Yueshenda 10’is as follows:Y=-9.791+0.356X_(12)+0.653X_(9)+3.027X_(2)+0.051X_(7)+0.468X_(10)+0.011X_(8),in which anthocyanins,vitamin C,single fruit dry weight,solid acid ratio,flavonoids and total sugar are important indicators for evaluating its quality.The principal component analysis method can be used to evaluate the fruit quality of‘Yueshenda 10’.This study provide a theoretical basis for the construction of quality evaluation system of‘Yueshenda 10’.

关 键 词：果桑 粤椹大10 主成分分析 品质评价 回归分析 

分 类 号：S663.9[农业科学—果树学]