Two-Stage Robust Security Constrained Unit Commitment Considering the Spatiotemporal Correlation of Uncertainty Prediction Error

Abstract

Robust optimization (RO) is an important tool to solve the security-constrained unit commitment (SCUC) problem for a power system with large-scale wind power. The main disadvantage of RO is that it is overly conservative, and the conservativeness of RO can be attributed to the uncertainty set used in the formulation. This paper proposes a two-stage robust SCUC model considering the spatiotemporal correlation of the uncertainty prediction error. First, based on the historical data, a polyhedral uncertainty set that can describe the spatial-temporal correlation of uncertainties is established, and the analytical relationship between confidence probability and the uncertain set is given. Second, a two-stage robust SCUC model with the objective of minimizing the operating cost under the forecasting scenarios is proposed based on the polyhedral set. Third, the Benders decomposition method is used to solve the proposed model according to its characteristics. The simulation results on the modified IEEE-30 and IEEE-118 bus system demonstrate that the proposed method can reduce the conservativeness of RO and guarantee the security and economy of the unit commitment.