基于总出力调度图与出力分配模型的梯级水电站优化调度规则研究

Study on the optimal operating rules for cascade hydropower stations based on output allocation model

本文基于大系统聚合分解理论,建立了梯级水电站总出力调度图(聚合)与出力分配模型(分解)相结合的双层优化模型,分别采用NSGA-Ⅱ多目标遗传算法和离散微分动态规划(DDDP)进行优化,得到了梯级水电站优化调度规则,可大大降低计算的复杂度。通过在出力分配模型中引入弃能项,有效地降低了电站的弃水,增加了梯级发电量。以清江梯级水电站为研究背景,优化调度规则在满足梯级保证出力的条件下,年均发电量较原设计方案可提高2.25亿kW.h,增长率达3.07%,减少弃水9.87亿m3,达35.83%,说明了模型和算法的有效性和合理性,为制定梯级水电站优化调度规则提供了新的思路。

In this paper, the two-layer optimal model of cascade hydropower station operation is established based on the theory of aggregation-decomposition of large-scale system with operation rule curves（aggregation） and output allocation model （decompostion）. The NSGA-Ⅱ multi-objective genetic algorithm and the discrete differential dynamic programming （DDDP） are used to solve the optimization model respectively. So the complexity of calculation is greatly reduced and the operating rules is derived basd on the principle of optimization-simulation. By introducing energy of spilling water to output allocation model, water spilled is greatly reduced and power generation is increased. The model and algorithms are effective and reasonable for the Qingjiang cascade reservoirs as a case study. Comparing with the designed scheme, the proposed method can generate extra about 2.25 ×10^8 s kW· h of electrical energy （by 3.07 % ） and save 9.87 ×10^8 m3 of water resources （by 35.83 % ） annually without reducing designed firm power. It is shown that the proposed method provides a new way to solving optimal operation problems for cascade reservoirs.