清江梯级水库群联合优化调度图研究

Derivation of the optimal operating curve rules for Qingjiang cascade reservoirs

针对清江梯级水库群中长期调度问题,根据最优调度结果确定了调度图的型式,引入了可能出力的概念,对水库群进行聚合,建立梯级总出力与梯级可能出力之间的关系.然后根据总出力是否为保证出力进行决策分解,由此采用一连串的二维水库调度图来描述水库群的联合调度规则.基于模拟调度,建立了优化模型,采用多目标遗传算法NSGA-Ⅱ进行优化计算,得到较好的联合优化调度图.模拟计算表明：对于1951-2005年实测径流系列,若采用梯级水库群联合优化调度图较常规调度方案年均发电量增幅可达2.62%;对于两种随机模拟方法生成的水文径流系列,若采用梯级水库群联合优化调度图,年均发电量可分别提高1.77%和2.52%,理论上经济效益显著.

Deriving the optimal operating curve rules for Qingjiang cascade reservoirs based on aggregated reservoir method was studied.The determined dynamic programming method was used to solve the optimization model.By analysis of the optimal operating tracks,the pattern of curve rules was obtained based on aggregated reservoir method.As the simulation-based model was built, a multi-objective genetic algorithm,NSGA-Ⅱ,was used to solve it and yielded a set of Pareto solutions.The Qingjiang cascade hydropower plants were selected as the case study,and there is an increase of 2.62 % in the hydropower generation by inputting the series during 1951 to 2005 period.As compared with the conventional rulse,the proposed method can increase hydropower generation by 1.77 % and 2.52 % respectively by using two types of synthetic hydrological data.