Identifying changing patterns of reservoir operating rules under various inflow alteration scenarios

Operating rules are important in the long-term operation of reservoirs for its capability of coping with inflow uncertainty. The characteristics of inflow vary as a result of climate change and human activities, and using stationary operating rules would lead to inefficient reservoir operation. This study focuses on identifying changing patterns of operating rules under various inflow alteration scenarios. Two simulation methods, the simple adjustment method (SAM) and the stochastic reconstruction method (SRM), are used to generate three inflow alteration scenarios: shifts of mean, coefficient of variation (CV), and seasonality. A deterministic reservoir optimization model is established and then resolved using discrete differential dynamic programming algorithm. Finally, the operating rules under each scenario are derived using the linear fitting method. China's Three Gorges Reservoir is used as a case study. The results show that the SAM and SRM produce similar operating rules, which are sensitive to inflow changes during refill and drawdown periods. It is shown that (1) the increase (decrease) of inflow mean changes the operating rules, resulting in the increase (decrease) of the water releases while the shift of CV has little impact on operating rules; (2) the seasonality changes operating rules in opposite directions during refill and drawdown periods; (3) the changing patterns of operating rules would be superimposed by the superposition of various inflow alteration scenarios whereas the effects might be not obvious. These findings are helpful for adaptive operation of reservoirs under changing environment.