Research on the Dispatching Decision Method of Cascade Hydropower Stations Based on the BVWS
摘要
Compared with the operation of a single hydropower station, the difficulty of operation and management for cascade hydropower stations increases exponentially. Especially during the critical periods such as concentrated drawdown before flood season and concentrated storage during late flood season, the dispatching strategies of cascade stations are crucial for successfully realizing operation objectives and enhancing power generation benefits of the cascade system. Scientific methods should be used to formulate long-, medium-, and short-term operation strategies, and the sequence of drawdown or storage for cascade reservoirs should be arranged reasonably to maximize the system benefits. Current research typically utilizes reservoir dispatching diagrams or optimal dispatching models to develop scheduling plans. Reservoir dispatching diagrams can be used to quickly access the operation schemes for reservoirs, but it cannot guarantee the optimal power generation benefits within a given dispatching period. Optimal models typically utilize algorithms such as dynamic programming to solve and obtain corresponding scheduling solutions, achieving the maximization of reservoir dispatching objectives. However, the problem of dimension disaster often easily occurs. To balance the optimization of system benefits and the efficiency of formulating operation plan, this paper proposes a benefit evaluation index, namely Benefit Variation from Water Storage (BVWS), which simultaneously couples water head benefits and backwater jacking influence of hydropower stations.Based on this, the graphs of BVWS for cascade stations are drawn, which can assist decision-makers in quickly formulating dispatching strategies for different time scales. Taking a cascade system composed of four hydropower stations in the upper reaches of the Yangtze River as the research object, this paper compares the calculation results and efficiency of the proposed method with the progressive optimality algorithm (POA). The results indicate that the proposed method can remarkably reduce the time required to develop optimal operation schemes, enabling a rational allocation of water resources among the cascade stations, therefore proving the scientific rationality of this method.