高级检索

    最优分辨系数和主成分分析在烤烟品种综合评价中的应用

    Application of Optimal Resolution Coefficient and Principal Component Analysis in Comprehensive Evaluation of Flue-cured Tobacco Varieties

    • 摘要: 为快速找出最适宜昆明生态条件的烤烟新品种,根据灰色系统理论,选择10 个烤烟品种的10 个性状作为评价因子,采用最优分辨系数确定了关联系数,采用主成分分析确定了性状的权重,建立了综合评价体系。结果表明,外观质量得分是所有烤烟品系中最重要的影响因子,权重达0.405。各品种综合得分依次为云烟100 > 云烟105 > PVH2254 > 云烟116 > NCT6 > NC196 > K326 > GL338 > NC71 > 云烟87。聚类分析表明,替代昆明地区云烟87 和K326 的最佳品种为云烟105,其次为PVH2254。云烟100 的综合评价得分最高,是特色化生产的最佳选择。

       

      Abstract: To quickly identify the most suitable tobacco varieties in Kunming, 10 characters of 10 flue-cured tobacco varieties were selected as evaluation factors. Optimal resolution coefficients of correlation coefficients were determined and principal component analysis was performed to identify trait weights to establish a comprehensive evaluation system according to the gray system theory. The results showed that the appearance quality score was the most important influence factor in all varieties and the weight was 0.405. The performance of all varieties were in the order of Yunyan100 > Yunyan105 > PVH2254 > Yunyan 116 > NCT6 > NC196 > K326 > GL338 > NC71 > Yunyan 87. Cluster analysis showed that the best alternative varieties of flue-cured tobacco was Yunyan 105, and PVH2254 followed. Yunyan 100 had the highest score of comprehensive evaluation and was the best choice for production characteristics.

       

    /

    返回文章
    返回