| 摘要: |
| 为制定满足渠系运行实际情况和作物生长发育所需土壤含水量条件的渠系优化配水方案，将考虑输水渗漏损失的渠系优化配水模型与土壤水量平衡模拟模型耦合，构建基于渠系输水模拟与土壤水量平衡模拟的两级渠系优化配水模型，使用改进的遗传算法求解多目标模型的Pareto解集。选择黑河流域中游西干灌区下的两级渠系作为研究区域，得到50组非劣的渠系优化配水方案。在Pareto前沿面上选取3组代表方案进行对比分析，再将3组中最优的方案与灌区实际灌水方案进行对比，结果表明:Pareto解集可以体现目标之间的博弈关系，当权重值向配水量最少目标偏移时会对土壤水含量等指标产生影响，进而产量下降；当权重值向产量最大目标偏移时会产生水资源浪费现象；而当目标函数的权重值处于中间水平时，产量和配水量目标都可以得到较好地满足，并且与实际灌水方案相比配水量减少113.54万m3、总产量增加660 t，能够达到节水增产的效果。该模型在兼顾渠系运行条件和土壤含水量的前提下权衡了2个目标之间的博弈关系，Pareto解集得出的结果能够满足决策者对于目标的不同需求，为灌区管理者的决策提供理论依据和指导。 |
| 关键词: 渠系优化配水 渠系输水模拟 土壤水量平衡 遗传算法 Pareto解集 |
| DOI：10.11841/j.issn.1007-4333.2020.05.16 |
| 投稿时间：2019-09-29 |
| 基金项目:国家重点研发计划课题(2016YFC0400207) |
|
| Irrigation water optimal allocation for two-level canals based on canal water transfer simulation and soil water balance simulation |
|
PAN Qi1,XIA Shuang2,GUO Shanshan1,SHAN Baoying1,GUO Ping1*
|
| (1.College of Water Resource & Civil Engineering, China Agricultural University, Beijing 100083, China;2. China Water & Power Press Company Limited, Beijing 100038, China) |
| Abstract: |
| In order to formulate an optimal canal water distribution schedule which satisfies the actual conditions of canal system operation and soil moisture demand for crop growth, this study coupled the canal water distribution optimization model and soil moisture simulation model to build a two-level canal water distribution optimization model based on canal water transfer simulation and soil water balance simulation. Improved genetic algorithm was used to solve the Pareto optimal set, two-level canal in Xigan irrigation district in the middle reaches of Heihe River basin was taken as the research area. A total of 50 non-inferior optimal canal water distribution schemes were obtained. Three groups of representative schemes were selected on Pareto frontier for comparative analysis. The best scheme was then compared with actual irrigation scheme. The results showed that Pareto optimal set could reflect the tradeoff between the targets. When the weight deviated to the minimum target of water distribution, it affected the soil water content and other indicators, and the yield would decrease. When the weight deviated to the maximum yield target, it would cause the water waste. However, when the weight was in the middle level, both the yield and water distribution targets were well met, and compare with the actual irrigation scheme, the amount of water decrease by 1. 135 4 million m3 and the total yield increase by 660 t. The weight in the middle level achieved the effect of saving water and increasing yield. In this model, the tradeoff between the two targets was based on both canal system operating conditions and soil water content, and the results of Pareto optimal set could satisfy the different demands of decision makers for targets. This study provided theoretical basis and guidance for the decision making of various managers in irrigation district. |
| Key words: optimal canal water resource allocation canal water transfer simulation soil water balance genetic algorithm Pareto optimal set |
