基于博弈论组合确权的模糊物元模型决策膜下滴灌向日葵调亏模式

Using Fuzzy Matter-element Model and Weighted Game Theory to Optimize Deficit Irrigation Scheduling for Sunflower

【目的】优选河西绿洲农区膜下滴灌向日葵适宜的水分调控策略。【方法】借助赋权的模糊物元模型,以大田向日葵为试验对象,对比评判向日葵产量、品质及水分生产率对不同水分调亏模式的响应结果。试验设置轻度(65%FC~75%FC,FC为田间持水率)、中度(55%FC~65%FC)和重度(45%FC~55%FC)3个亏水梯度,在向日葵苗期与成熟期实施不同程度调亏组合,剩余时期保持充分供水,共计6个处理,而全生育期充分(85%FC~75%FC)供水作为对照处理,各处理3次重复,采用完全随机区组试验布置。【结果】水分调亏对向日葵各指标的影响不同,建立了向日葵形态、产量、品质及水资源利用4个方面9个指标的多层次综合评价体系,采用博弈论组合赋权得出2019年株高权重最大,为0.141,粗蛋白量最小,为0.083,2020年权重最大是粗脂肪,为0.162,最小是水分生产率,为0.065;基于模糊物元耦合模型进行综合评判,连续2个生长季综合得分排序一致,其中苗期轻度或中度且成熟期轻度处理的欧氏贴近度相对最大。【结论】来水充裕时,苗期轻度且成熟期轻度调亏模式可获得较高的籽粒产量和品质;来水不足时,苗期中度且成熟期轻度调亏模式是较佳的节水灌溉模式。

【Objective】Sunflower is a cash crop grown in arid northwestern China.Improving its water use efficiency plays an important role in developing water-saving agriculture in that region.The aim of this paper is to find an optimal regulated and mulched deficit drip irrigation scheduling for it.【Method】The experiment was conducted in 2019—2020 in a sunflower field at Hexi Oasis.Water deficit was imposed at seedling and maturity stages,with soil water in other growing stages kept sufficient for roots to extract.Overall,there were six treatments,with the water deficit induced by keeping the soil water content at 75%~65%(mild),65%~55%(moderate)and 55%~45%(severe)of the field capacity,respectively.The control was sufficient irrigation with the soil water content kept at 75%~85%of the field capacity.Each treatment had three replicas,which were designed using the completely random block experiment arrangement.The experimental results were analyzed using an improved fuzzy matter-element model coupled with the weighted game theory to quantify the effect of different water deficits on yield and quality,as well as water productivity of the crop.【Result】The effect of water deficit on the crop traits varied.A multi-level evaluation system revealed that in 2019,crop height had the highest weight(0.141)and the crude protein content had the least(0.083),while in 2020,crude fat content had the highest weight(0.162)and water use efficiency had the least(0.065).Comprehensive evaluation of the fuzzy matter-element model showed that the Euclidean degree was the best for a combination of mild or moderate water deficit at seedling stage and mild water deficit at maturity stage.【Conclusion】When water is sufficient,a mild regulated deficit irrigation(RDI)at the seedling stage coupled with a mild RDI at the maturity stage produces the highest seed yield,while when water is insufficient,a moderate RDI at the seedling stage combined with a mild RDI at the maturity stage works the best.The proposed model for optimizing regulated deficit irrigation is applicable not only for sunflower watered by mulched drip irrigation,but also for other crops irrigated by other methods.