Abstract: To quantify the peak-cutting of flood control storage capacity of the Three Gorges Reservoir (TGR), an optimization-fitting framework was employed to calculate the historical storage flood data. Various maximum water level schemes during scheduling were set at first, and a dynamic planning algorithm within the framework was utilized to optimize and solve different incoming floods. By doing so, the flood peak reduction curves under different flood control storage capacities were obtained. Furthermore, regression analysis was employed to quantify the peak-cutting of flood control storage capacity of TGR. Results demonstrate that the correlation between the utilization of the flood control storage capacity in TGR and the reduction of flood peak flow can be classified into three types: “flat curve following an increment”, “segmented increment”, and “linear increment”. The first two types exhibit inflection points, suggesting that the flood control storage capacity should be maintained below a specific threshold to fully maximize its efficient peak-cutting capacity. As for the third type, the flood control storage capacity can be planned rationally for dispatching in line with the flood peak reduction target. The findings provide valuable insights for the scientific and rational allocation of flood control storage capacity in the Three Gorges Reservoir.

Key words: Three Gorges Reservoir, flood control benefits, flood control dispatch, dynamic programming, flood peak attenuation capacity