基于运行数据的水电站MILP模型最优代表水头选择取方法

An Optimal Head Selection Method for MILP Moder of Hydropower Plant Based on Operation Data

摘要:

针对采用设计水头的水电站混合整数线性规划（mixed integer linear programming, MILP）调度模型计算的出库流量与实际出库流量偏差较大的问题，提出了基于运行数据的水电站MILP模型最优代表水头选取方法。首先基于运行数据，采用MILP模型拟合出使模型计算出库流量过程与水电站实际出库流量过程偏差最小的代表水头，然后在实际调度中，以日平均入库流量和日平均出力作为该代表水头特征向量，根据预测入库流量和日计划电量即可选取最优代表水头。计算实例表明，相比于传统固定水头，该方法能够更好的反应水电站实际的出库过程，有利于提高电网制定调度计划中梯级水电站上下游水量匹配精度。

Abstract:

In order to solve the deviation between the actual discharge and the calculated discharge of hydropower plant by mixed integer linear programming (MILP) model with design head, an optimal head selection method for MILP model of hydropower plant based on operation data is proposed. Firstly, the MILP model of hydropower plant is used to obtain the optimal head which has the minimum deviation between the actual discharge and the calculated discharge. Then, the daily average reservoir inflow and the daily average output are taken as the optimal head eigenvectors. In the actual scheduling process, the optimal head can be selected according to the predicted reservoir inflow and the daily planned energy. The example shows that the proposed method can better reflect the actual discharge of hydropower plant, and is conducive to improving the matching accuracy of upstream and downstream water volume of cascaded hydropower plants in the formulation of power grid dispatching plan.