Dual time-scale robust optimization for energy management of distributed energy community considering source-load uncertainty

The uncertainty of the renewable energy output and load demand poses significant challenges to the efficiency and stability of energy community operation. To reduce the adverse effects of uncertainty factors, this paper proposes a dual time-scale energy management method for a distributed energy community. In the day-ahead scheduling, an adaptive robust optimization model is proposed. This model uses the Dirichlet process mixture model (DPMM) to establish the fuzzy set of uncertainty for solar power generation and load demand. The column-and-constraint generation algorithm (C&CG) to solve the optimization model. In the intra-day scheduling, a rolling optimization method based on model predictive control (MPC) is combined with short-term source-load forecasting, which ensures consistency between the intra-day and day-ahead scheduling plans and real-time supply-demand balance. Experiments show that the robustness of energy management in the proposed method is enhanced significantly in various scenarios. At the same time, the operating costs of energy community are also reduced with different degrees of conservatism.