基于灰色关联度分析和DTOPSIS法综合评价小麦新品系在云南省的适应性

金轻 赵红 林丽萍 杨忠慧 李宏生 刘琨 杨木军 李绍祥 丁明亮

摘要：【目的】利用灰色關联度分析和DTOPSIS法综合评价小麦新品系在云南省的适应性，为小麦品系综合评价及其在云南省示范推广提供理论依据。【方法】采用灰色关联度分析及基于灰色关联度分析的DTOPSIS法（以下简称为DTOPSIS法）对2018─2019年度云南省区域试验中11个小麦新品系在云南省的适应性进行综合评价，比较小麦品系不同评价方法的准确性，并筛选出综合表现较好的小麦品系。【结果】11个小麦新品系产量排序为云152-484>云麦110>临麦22>玉18-2>云麦56>蜀麦1767>德1733>云184-13>文2-396>滇麦13号>保15J-8>云杂19号，超过对照品系云麦56的品系共有4个，分别为云152-484、云麦110、临麦22和玉18-2。8个农艺性状的权重值排序为基本苗>产量>最高分蘖数>有效穗数>穗粒数>千粒重>株高>生育期。灰色关联度分析结果显示，供试小麦品系与理想品系的关联度（Gi）排序为云152-484>玉18-2>云麦110>云麦56>滇麦13号>保15J-8>蜀麦1767>德1733>文2-396>临麦22>云杂19号>云184-13。DTOPSIS法分析结果显示，供试小麦品系与理想解的Ci排序为云152-484>云麦110>玉18-2>云麦56>蜀麦1767>滇麦13号>德1733>临麦22>文2-396>保15J-8>云184-13>云杂19号。虽然2种方法评价结果整体趋势较相似，但也存在差异，与按产量表现的排名结果相比，按DTOPSIS法的排名结果较灰色关联度分析的排名结果更吻合。此外，Ci差异大于Gi差异和产量差异，且按DTOPSIS法Ci排名与按产量表现排名和按灰色关联度分析Gi排名均呈极显著正相关（r=0.811和r=0.895，P<0.01），进一步说明DTOPSIS法较灰色关联度分析和产量表现评价小麦新品系在云南省的适应性更充分、更合理。【结论】综合评价云南小麦品系时不仅要重视品系的产量，还应重视小麦的基本苗、最高分蘖和有效穗等性状的考察。云152-484、云麦1102和玉18-2在云南省适应性和丰产性均较好，可进一步试验、示范及推广。基于灰色关联度分析的DTOPSIS法更适于综合评价云南省小麦新品系的适应性。

关键词： 小麦;灰色关联度分析;DTOPSIS法;适应性;综合评价

中图分类号： S512.102.4                            文献标志码： A 文章编号：2095-1191（2020）10-2440-07

Comprehensive evaluation of adaptability of new wheat lines in Yunnan based on grey relevance analysis and DTOPSIS method

JIN Qing1， ZHAO Hong 2，3， LIN Li-ping4， YANG Zhong-hui 2，3， LI Hong-sheng 2，3，

LIU Kun 2，3， YANG Mu-jun 2，3， LI Shao-xiang 2，3*， DING Ming-liang 2，3*

（1Zhaotong Academy of Agriculture Sciences， Zhaotong， Yunnan  657000， China; 2Food Crops Resource Institute，Yunnan Academy of Agriculture Sciences， Kunming  650205， China; 3Yunnan Branch of National Wheat Improvement Center， Kunming  650205， China; 4Seed Management Department of Yunnan， Kunming  650031， China）

Abstract：【Objective】The purpose was to provide theoretical basis for comprehensive evaluation of wheat lines and their demonstration and popularization in Yunnan. The adaptability of new wheat lines in Yunnan was comprehensively evaluated by grey correlation analysis and DTOPSIS method. 【Method】The adaptability of 11 new wheat lines in the regional trial of Yunnan from 2018 to 2019 were analyzed by using grey correlation analysis method and DTOPSIS method based on grey correlation analysis. The accuracy of different methods was compared， and wheat lines with better comprehensive performance were selected. 【Result】The yield of 11 new wheat lines ranked from high to low as Yun 152-484>Yunmai 110>Linmai 22>Yu 18-2>Yunmai 56>Shumai 1767>De 1733>Yun 184-13>Wen 2-396>Dianmai 13>Bao 15J-8> Yunza 19. The yields of four new wheat lines were higher than that of the control variety， they were Yun152-484， Yunmai110， Linmai22 and Yu18-2. The weight of eight traits of new wheat lines from high to low was basic seedling>yield per unit area>maximum tiller number>effective panicles>grains per panicle>thousand-grain weight>plant height>growth period. The results of grey correlation analysis showed that the Gi values between the tested materials and ideal variety ranked from high to low as Yun 152-484>Yu 18-2>Yunmai 110>Yunmai 56>Dianmai 13>Bao 15J-8>Shumai 1767>De 1733>Wen  2-396>Linmai 22>Yunza 19>Yun 184-13. The results of DTOPSIS analysis showed that the Ci values of the tested materials and ideal variety ranked from high to low as Yun152-484>Yunmai110>Yu 18-2>Yunmai 56>Shumai 1767>Dianmai 13>De 1733>Linmai 22>Wen 2-396>Bao 15J-8>Yun 184-13>Yunza 19. It could be seen that although the overall trend of the two methods was similar， but there were also differences. Combined with the yield performance of each line， the ranking result of DTOPSIS Method was better than that of grey correlation analysis method.In addition，the difference of Ci value was greater than that of Gi value and yield. Moreover， Ci value ranking was extremely positively correlated with yield performance ranking and Gi value ranking（r=0.811 and r=0.895，P<0.01）. The results further showed that the DTOPSIS method based on the grey correlation analysis was more sufficient and reasonable than the gray correlation analysis and yield performance to evaluate the adaptability of new wheat lines in Yunnan. 【Conclusion】In Yunnan， the comprehensive evaluation of new wheat lines should not only pay attention to its yield performance， but also its basic seedling， maximum tiller number， effective panicles and other agronomic traits. The DTOPSIS method based on the grey correlation analysis is more suitable for comprehensive evaluation of adaptability of new wheat lines in Yunnan. Yun 152-484， Yunmai 110 and Yu 18-2 have better adaptability and high yield， which can be further tested， demonstrated and promoted in different ecological regions of Yunnan.

Key words： wheat; grey relevance analysis; DTOPSIS method; adaptability; comprehensive evaluation

Foundation item： National Key Research and Development Program of China（2016YFD0101603）

0 引言

【研究意义】小麦是我国第三大粮食作物，其可持续发展为保障我国粮食安全做出了重要贡献（何中虎等，2018）。小麦也是云南省的主要粮食作物，其产量约占云南粮食总量的10%（黄兴奇，2005;于亚雄，2016）。据统计，云南省小麦种植面积在20世纪90年代基本稳定在6.6……