基于RS、GIS及智能算法的渠系优化配水  

Optimal irrigation water distribution based on RS,GIS and intelligent algorithms

作　　者：刘照[1,2] 华庆伟 张成才[4] 丁小辉[1,2] 张树清[1] 程帅[1,2] 

机构地区：[1]中国科学院东北地理与农业生态研究所,吉林长春130102 [2]中国科学院大学,北京100049 [3]平顶山学院图书馆,河南平顶山467000 [4]郑州大学水利与环境学院,河南郑州450001

出　　处：《西北农林科技大学学报（自然科学版）》2017年第4期213-222,229,共11页Journal of Northwest A&F University(Natural Science Edition)

基　　金：国家重点研发计划"全空间信息系统与智能设施管理"项目;"全空间信息系统建模理论"课题(2016YFB0502301);河南省基础研究计划项目(152300410044)

摘　　要：【目的】基于RS、GIS技术及智能算法,对灌区渠系优化配水工作进行研究,为提高灌区管理、制定配水计划提供科学依据。【方法】采用"组间续灌、组内轮灌"的工作方式,选择总配水时间最短与轮灌组之间引水持续时间差异值同时最小作为优化目标,以黑河中游甘肃省张掖市甘州区盈科灌区盈二支一分支为例,构建多目标渠系优化配水模型,并基于粒子群算法和蚁群算法对模型进行求解。【结果】成功构建了多目标渠系优化配水模型,以黑河中游张掖市甘州区盈科灌区5km×5km区域优化配水为实例的研究表明,采用多目标粒子群算法时的最大配水时间为113.6h,最小配水时间为105.3h,轮灌组引水时间差异为8.3h;采用多目标蚁群算法时的最大配水时间为113.5h,最小配水时间为105.3h,轮灌组引水时间差异为8.2h。解算的结果均低于人工制定的168h原配水计划。【结论】基于RS、GIS及智能算法解决渠系配水问题,不仅可以缩短轮灌周期,减少配水过程中的水量损失,还可保证配水精度,减少劳动力支出。[Objective] This study investigated optimal irrigation water distribution in irrigation district based on RS,GIS and intelligent algorithms to improve irrigation management and provide scientific basis for making water distribution plans. [Method] A multi-objective optimization model of rotation irrigation was formulated to minimize the total irrigation duration of water distribution and the time difference among groups simultaneously. Taking Yingke second branch canal in Ganzhou, Zhangye on the middle reach of Heihe River as research object,the optimal model was instantiated and solved by particle swarm algorithm and ant colony algorithm. [Result] The multi-objective optimal irrigation water distribution model was for- mulated successfully. Taking Yingke 5 km×5 km region in Ganzhou, Zhangye on the middle reach of Heihe River as the optimal water distribution research area, multi-objective particle swarm algorithm showed that the maximum irrigation duration of water distribution was 113.6 h, the minimum irrigation duration was 105.3 h,and the time difference of rotational irrigation groups was 8.3 h. The multi-objective ant colony algorithm showed that the maximum irrigation duration of water distribution was 113.5 h,the minimum irri- gation duration was 105.3 h, and the time difference among rotational irrigation groups was 8.2 h, Both were less than 168 h based on traditional artificial method. [Conclusion] Adopting RS,GIS and intelligent algorithms in irrigation water delivery not only reduced the whole irrigation duration and water loss but al- so ensured accuracy of water delivery and reduced labor force.

关 键 词：渠系优化 智能算法 轮灌组合 RS与GIS 

分 类 号：S274.3[农业科学—农业水土工程]