沅水梯级水库组合动态约束条件优化调度研究

Optimal Operation of Yuanshui’s Cascade Reservoirs under Combined Dynamic Constraints

采用变状态空间增量逐次逼近轮库迭代动态规划优化方法,根据沅水梯级水库电站水资源侧、电源侧的目标层及动态约束条件实际,通过水库电站层、机组层、设备运行层的组合约束条件结构,对三板溪、白市、托口等沅水流域梯级水库电站,选取具有丰、平、枯典型代表的来水年组进行中长期联合优化计算,实现梯级水库电站优化调度应用,与实际运行进行指标比较,验证沅水流域梯级优化调度研究的应用价值。在沅水流域梯级水库优化调度研究中,进一步对梯级水库水力联系中的调度对象目标,构建算法组合计算模型、配置约束条件与目标相互转化要素,有效结合径流预报、优化调度、厂内机组设备高效率区调控经济运行,拓展长中短期优化调度应用。

In this paper, an iterative dynamic programming optimization method of variable state space increment successive approximation for reservoir-by-reservoir is adopted. According to the actual target layer and dynamic constraints of water, resources side and power supply side of Yuanshui cascade reservoirs(Sanbanxi, Baishi and Tuokou reservoirs) are treated by the combined constraints’ structure of reservoir power station layer, unit layer and equipment operation layer. The cascade reservoir power stations are selected to carry out mid-term and long-term joint optimization calculation with typical representative water coming years of wet, normal and dry water. The application of cascade reservoir power station optimal dispatching is realized. The optimal dispatching results are verified by comparing with the actual operation records. The target of operation in cascade reservoir hydraulic connection is further studied, and the algorithm combination calculation model, allocation constraints and target transformation elements are constructed. Effective combination of runoff forecasting, optimal dispatching and efficient area of plant unit equipment aims to regulate economic operation, expand the application of long-term, medium-term and short-term optimal dispatching.