Security-Constrained Multi-Stage Robust Dynamic Economic Dispatch Model with Bulk Storage

As wind penetration rates continue to increase, the main challenge faced by operators is how to schedule flexible resources in the future period to ensure the safe and stable operation of power grids under multiple uncertainties. In this study, we propose a security-constrained multi-stage robust economic dispatch model with storage (SMREDS) to address multiple uncertainties, including wind power outputs and N-1 generator contingency. Compared to the traditional two-stage robust dynamic economic dispatch model, the proposed multi-stage dispatch model yields sequential operation decisions with the uncertainties revealed gradually over time. To overcome the computational issue of multi-stage problems, with large-scale post-contingency constraints and the convergence issue of stochastic dual dynamic programming (SDDP) algorithm, this study also proposes a combined two-stage benders decomposition and robust dual dynamic programming (RDDP) to solve the proposed model. Compared to the traditional SDDP algorithm, the proposed RDDP algorithm uses the inner approximation and outer approximation methods to approximate the upper and lower bounds of the future cost-to-go function, which overcomes convergence issue of the traditional SDDP algorithm. We tested the proposed algorithm on the IEEE-3 bus, IEEE-118 bus, and the German power system. The simulations result verify the effectiveness of the proposed model and proposed algorithm