Impacts of Inflow Variations on the Long Term Operation of a Multi-Hydropower-Reservoir System and a Strategy for Determining the Adaptable Operation Rule

Obvious inflow variations resulting from changing environments bring big challenges to the operations of hydropower reservoirs. This study reveals the impacts of average annual inflow volume (AAIV) variations on the long term operation of a multi-hydropower-reservoir (MHR) system, and presents a strategy for determining the adaptable operation rule. The strategy includes two parts. One part is making different inflow scenarios based on the change points of AAIVs. Another part is applying the principle of cross validation to select the adaptable rule from the formulated operation rules in various inflow scenarios. Specifically, the change points of AAIVs are identified by three statistical methods. An optimization operation model of an MHR system is built, and three evolutionary and meta-heuristic algorithms are applied to resolve the model in different inflow scenarios. Based on the optimal operation results, two machine learning algorithms are employed to formulate operation rules in each inflow scenario. The MHR system at the upstream of Yellow River basin is taken as a case study. The results show that (1) the long term operation of an MHR system is sensitive to the AAIV variations; and (2) the presented strategy is feasible in determining the adaptable operation rule for an MHR system under the AAIV variations. The findings of the study are helpful for the long term operation of an MHR system under the AAIV variations.