南水北调东线工程江苏段多目标优化调度研究

Study on multi-objective optimal operation of Jiangsu Section of South-to-North Water Transfer Project

考虑南水北调东线工程供水目标、经济成本以及系统稳定性特点,本文以受水区生活、工业以及农业用水供水量最大,梯级总抽水量最小以及调水峰值最小5个目标为优化目标,构建了南水北调东线工程江苏段水资源优化调度模型。在此基础上,针对传统多目标蛙跳算法(MOSFLA)存在多样性差,后期易陷于局部最优解等缺点,本文提出了一种改进的多目标蛙跳算法(MQSFLA),通过引入实数量子编码、采用改进的量子旋转门局部更新策略以及基于动态外部档案集维护Pareto非劣解的方式,改善传统MOSFLA。分别基于MQSFLA和MOSFLA求解3种典型来水条件下的多目标优化模型,分析不同目标之间的关系,提出调蓄湖泊群联合优化调度和水量调配方案。结果表明:MQSFLA在多样性和收敛性方面都优于传统的MOSFLA,能得到更多高质量的非劣解集,更有效协调供水以及抽水的矛盾,满足不同来水条件下多水源相互补给的多目标优化调度,为南水北调东线工程江苏段的运行管理提供理论依据和科学支撑。

Considering the water supply,economic cost and system stability of Jiangsu Section of South-to-North Water Transfer(E-SNWT)Project,this paper firstly develops an optimal operation model of E-SNWT Project with minimizing the total pumpage,maximizing the domestic water supply,industrial water supply and agricultural water,and minimizing the diversion peak as the objective functions.To overcome the shortcoming of poor diversity and easily falling into the local optimal solution of original Multi-objective Shuffled Frog Leaping Algorithm(MOSFLA),this paper proposed Multi-objective Quantum Shuffled Frog Leaping Algorithm(MQSFLA)based on real quantum coding,modified quantum gate updating strategy and dynamic external archive set.The model was solved by MQSFLA to develop an optimal operation scheme for E-SNWT Project.Compared with original MOSFLA,it shows that the MQSFLA performs is better both in diversity and convergence,more effective coordination of water supply and pumping under different scenarios,which can provide theoretical basis and scientific support for the operation management of E-SNWT Project.