Optimization of water supply reservoir scheduling with insufficient regulating capacity

To make use of a limited regulating reservoir capacity to provide a small amount of water supply benefits, a multiobjective optimal model that considers the water demand and regulating capacity of the pumping station was established, by combining the maximum sequence similarity (F1), maximum pumping station suitability (F2), maximum ecological flow satisfaction (F3) and maximum number of continuous water supply periods (F4), which were solved by NSGA-II algorithm, K-means clustering, and EW-TOPSIS method were used to select the optimal scheme. The results showed that the results of the combination of F1 and F2 were better than that of the list method, with a multiannual average water supply of 3.88 × 107 m3; the occupation of ecological flow significantly increased the water supply efficiency, with a multiannual average water supply of 4.49 × 107 m3, with a multiannual average ecological satisfaction degree of 10.84; the interruption periods were gathered between January and May by accepting the interruption periods regularly occurred, the water supply was 1.10 × 107 m3 with 95% guarantee rate by taking June–December as the water supply period. Our data indicated that reservoirs with insufficient regulation capacity can produce certain water supply benefits, and there were competitive relationships between objectives, and competitiveness increases with the increase of pumping station regulation capacity.