灰色关联度分析和DTOPSIS法在贵州威宁引进苹果综合评价中的应用

【目的】建立苹果品质综合评价模型,筛选品质优良的品种(系),为贵州省威宁县苹果品种结构优化调整提供参考依据。【方法】测定引进贵州省威宁县种植的10个苹果品种(系)(编号Z1~Z10)果实的单果重(T1)、果形指数(T2)、去皮硬度(T3)、可溶性固形物(T4)、可滴定酸(T5)、固酸比(T6)和果实色泽L*(T7)、a*(T8)、b*(T9)]9项指标,利用灰色关联度分析联合DTOPSIS法对其进行综合评价,建立苹果品质综合评价模型。【结果】10个苹果品种(系)中,单果重、去皮硬度和固酸比指标最小均为Z9(16-112),但其可滴定酸含量最高,为0.60%;单果重、果形指数、去皮硬度、可溶性固形物含量、固酸比和L*值最大分别为Z3(2-14)、Z1(东71-31)、Z5(秦脆)、Z8(8-31)、Z5(秦脆)和Z9(16-112)。9项果实品质指标的权重值排序为可溶性固形物>去皮硬度>固酸比>果形指数>单果重=a*>可滴定酸>L*=b*。灰色关联度分析结果显示,灰色评判值(G_i)排序为:Z5(秦脆)>Z2(南39-66)>Z1(东71-31)>Z4(147)>Z7(39-134)>Z8(8-31)>Z10(92)>Z3(2-14)>Z6(秦蜜)>Z9(16-112)。DTOPSIS法分析结果显示,与理想品种接近度(C_i)排序为:Z5>Z2>Z1>Z10>Z3>Z7>Z4>Z6>Z8>Z9。2种评价方法排名趋势线吻合度较高,在分析过程中均涵盖9项果实品质指标信息,具有可靠的统计学基础,其中C_i值差异大于G_i值差异;通过2种方法的比较建立苹果品质评价预测模型:Y=0.09T1+0.11T2+0.17T3+0.19T4+0.08T5+0.15T6+0.06T7+0.09T8+0.06T9。【结论】DTOPSIS法对苹果品质的综合评价效果较好,通过权重系数建立的评价模型可用于苹果品质的综合评价,灰色关联度分析联合DTOPSIS法评价结果排序前三均为秦脆、南39-66和东71-31,可作为优化项目区苹果品种结构的重点品种(系)。

Application of grey correlation analysis and DTOPSIS method incomprehensive evaluation of introduced apples in Weining,Guizhou

【Objective】In order to establish the comprehensive evaluation model of apple quality and screen out the varieties(lines) with good quality,and provide theoretical basis for the optimization and adjustment of apple variety structure in Weining,Guizhou.【Method】Nine indexes including fruit weight(T1),fruit shape index(T2),peeling hardness(T3),soluble solids(T4),titratable acid(T5),the ratio of solid and acid(T6) as well as fruit color including L*(T7),a*(T8) and b*(T9) of 10 apple varieties(lines) (numbers were Z1-Z10) introduced to Weining,Guizhou were determined.Grey relational degree analysis and DTOPSIS method were used to evaluate apple qualitycomprehensively,and a comprehensive evaluation model was established.【Result】Among the 10 apple varieties(lines),Z9(16-112) had the lowest indexes of fruit weight,peeling hardness and solid acid ratio,while it had the highest value of the titratable acid content which was 0.60%. Z3(2-14),Z1(East 71-31),Z5(Qincui),Z8(8-31),Z5(Qincui) and Z9(16-112) had the highest fruit weight,fruit shape index,flesh hardness,soluble solid content,solid acid ratio and L* value,respectively. The weight values of nine fruit quality indexes were as follows:soluble solid>peeling hardness>solid acid ratio>fruit shape index>fruit weight=a*> titratable acid >L*=b*. The results of grey relational analysis showed that the G_i values were Z5(Qincui) >Z2(South 39-66) >Z1(East 71-31) >Z4(147) >Z7(39-134) >Z8(8-31) >Z10(92) >Z3(2-14) >Z6(Qinmi) >Z9(16-112). The proximity to ideal variety(C_i) values of DTOPSIS analysis were Z5>Z2>Z1>Z10>Z3>Z7>Z4>Z6>Z8>Z9. The ranking trend lines of the two evaluation methods were highly consistent. The two evaluation methods both covered most of the information of fruit quality indicators in the analysis process,which had a reliable statistical basis. The difference of C_i values was greater than that of G_i values. Apple quality evaluation model Y=0.09T1+0.11T2+0.17T3+0.19T4+0.08T5+0.15T6+0.06T7+0.09T8+0.06T9 was established combinedgrey relational degree analysis with DTOPSIS.【Conclusion】The comprehensive evaluation effect of DTOPSIS method is better,and the evaluation model established by the weight coefficient can be used for the comprehensive evaluation of apple quality. The top three evaluation results of grey relational degree analysis combined with DTOPSIS method are Qincui,South 39-66,and East 71-31,which can be used as the key varieties(lines) to optimize the structure of apple varieties in the project area.