基于自优化模拟技术的梯级水电站耗能最小模型

Minimum Energy Consumption Model of Cascade Hydropower Stations Based on Self-Optimization Simulation Technology

摘要: 针对水库优化调度模型求解方法中，传统动态规划(DP)法存在“维数灾”，求解耗时较长等不足，提出基于自优化模拟技术的水电站水库调度模型快速求解方法(SOST)。该方法在满足水电站负荷要求和最小下泄流量要求的前提下，尽可能提高水库调度期的运行水位以寻求全局优化调度解。选取潘口–小漩梯级水电站30个典型日的历史运行数据作为计算实例，汛期典型日的小时尺度优化调度结果显示：该方法能够实现潘口–小漩梯级水电站总负荷的经济分配，优化分配计算所得梯级水电站总耗能为104.23万kW•h，比实际运行耗能减少3.89%。相较于传统的DP算法，该方法计算时间降幅达58.04%。SOST法寻优性能较好，可有效提高水能利用效率，为梯级水电站发电优化调度提供参考。

Abstract: The traditional dynamic programming (DP) method has the disadvantages of “dimension disaster” and long solving time. According to the characteristics of low water consumption rate of reservoir in high head operation, a fast solution method for reservoir operation model of hydropower station based on self-optimization simulation technology (SOST) is proposed. Under the premise of meeting the load requirements and minimum discharge requirements of hydropower station, the method can improve the operation level of reservoir as much as possible to seek the global optimal operation solution. Taking the historical operation data of 30 typical days of Pankou-Xiaoxuan cascade hydropower sta-tions as an example, the hourly optimal operation results of typical days in flood season show that this method can realize the economic distribution of total load of Pankou-Xiaoxuan cascade hydropower stations, and the total energy consumption of the cascade hydropower stations calculated by the op-timal distribution is 1,042,300 kW•h, which is 3.89% less than the actual operation energy consump-tion. Compared with the traditional DP algorithm, the calculation time of this method is reduced by 58.04%. The results show that the SOST method has good performance and can effectively improve the utilization efficiency of water energy, which provides a reference for the optimal operation of cascade hydropower stations.