SSDP Model with Inflow Clustering for Hydropower System Operation

Sampling stochastic dynamic programming (SSDP), which considers the uncertainty of streamflow, is a popular and useful method for solving release decisions of reservoirs. It is easy to implement the long-term operation for cascaded hydropower system with poor inflow prediction ability. Furthermore, SSDP describes the temporal and spatial structure of the stochastic streamflow processes implicitly through inflow scenarios instead of representing the multivariate distribution of inflow by conditional probabilities in stochastic dynamic programming (SDP). However, computation time of SSDP procedure will increase exponentially with the growth inflow scenarios. Thus, clustering algorithm is employed to reduce the number of inflow scenarios in order to improve the efficiency and operability of SSDP in practical applications. The calculation results of SSDP with inflow clustering are analyzed with different cluster numbers. The principle of how to find the least inflow scenarios to represent all inflow sequences has also been proposed. The least inflow scenarios and relevant probabilities found by clustering algorithm can approximate the empirical distribution of all streamflow scenarios used in this study without obviously decreasing the energy and exacerbating the shortage of firm power.