Optimizing Water Resources Allocation and Hydropower Generation for Supporting Reservoir Management

In this paper, a multi-stage fuzzy stochastic programming (MFSP) method is developed to optimize water resources allocation under uncertainties. MFSP method can not only deal with uncertainties expressed as fuzzy sets and probabilistic distributions, but also provide a linkage between the pre-regulated policies and the associated economic implications. Four hydropower generation targets, three inflow levels and five violation risk levels are analyzed. Results indicate that: (i) system benefit would reduce among with the rise of $\alpha$ and $\beta$ levels, which range from 4.16 to 5.08 × 109 US$; (ii) since agriculture is still the largest water user in the future, it is desired to reduce agricultural water consumption through adjusting crop planting structure and water allocation scheme; (iii) inflow levels have significant influence on water allocation pattern. With the growth of inflows, the total allocated water would increase 8.34 × 109m3. Therefore, managers should pay emphasis on the reservoir management in different inflows and consider storing