Analysis of factors affecting flood control and power generation under reservoir rolling forecasting optimized operation

Abstract

[Objective] Reservoir scheduling is an important non engineering measure for the comprehensive utilization of water resources at present. In recent years, with the improvement of hydrological forecasting technology, the reservoir optimal operation combined with hydrological forecasting has received increasing attention. However, the impact mechanism of flood control and power generation benefits under the reservoir rolling forecasting optimized operation is still unclear. [Methods] In response to this issue, a rolling forecast optimization scheduling model for reservoir flood control was established, and the control variable method was used to analyze the impact of different flood levels, forecast periods, and dynamic control upper limits of flood season water levels on reservoir flood control and power generation benefits taking Xiajiang Reservoir as a case. [Results] The results show that:(1) The peak shaving rate of floods gradually decreases with the increase of the upper limit of dynamic control of water level during flood season.(2) The power generation of the reservoir increases with the increase of the upper limit of the dynamic control of water level during the flood season, and the maximum discharge flow also increases.(3) The larger the magnitude of the flood, the shorter the forecast period required for reservoir operation to achieve maximum peak shaving effect.(4) The difference in the optimization scheduling result of flood control rolling forecasting under uncertain and deterministic inflow conditions is relatively small. [Conclusion] In summary, under the reservoir rolling forecasting optimized operation, there are patterns in the impact of flood magnitude, forecast period, and dynamic control upper limit of flood season water level on flood control and power generation. Combined with reliable forecast information, raising the flood season water level limit of the reservoir can improve power generation efficiency under the premise of controllable risk. Taking a 50-year design flood and the forecast period is 72 hours as an example, compared with the upper limit of dynamic control of water level during flood season being 46 m, when the upper limit is set at 43.5 m, the average peak shaving rate only increases by 0.46%(about 104 m³/s), but the average power generation decreases by 30.55%(about 15.56 million kWh).

Keywords

rolling forecast optimization scheduling / flood control scheduling / power generation scheduling / flood prediction period / flood control water level / flood forecasting / discharge / numerical simulation