Adaptive Robust Optimization Based on Data-driven Uncertainties of Source and Load for Microgrid Operation

Abstract

The strong uncertainty of renewable energy output and load demand makes the stable operation of microgrids a challenging and important issue. However, the scheduling methods based on deterministic models cannot accurately describe the influence of uncertainties on operation of microgrids. To address this problem, this paper proposes a two-stage adaptive robust optimal scheduling model (TSARO) that considers both source and load uncertainties. The model first adopts the dirichlet process mixture model (DPMM) to perform cluster analysis and parameter estimation on massive historical data, and constructs a data-driven uncertainty set of source and load. Then, based on the uncertainty set, the TSARO model aiming at minimizing the microgrid operation cost is developed under the worst-case scenario. Finally, this paper solves the optimization model using column constraint generation algorithm (C&CG) to obtain the day-ahead power dispatching plan. Simulation results show that the proposed model in this paper has better economic benefits compared with several classical optimization models.