One of the requirements for a resilient distribution system with distributed generators (DGs) is to restore as many critical loads as possible during a power supply outage of the main power grid. During the outage period, the distribution network is divided into microgrids (MGs) and continuously provides power to loads. However, the uncertainty of renewable DGs’ output, which cannot be predicted perfectly, increases the difficulty of optimal operation of MGs. In this paper, an MG scheduling method is proposed based on robust model predictive control (RMPC) for the resilience enhancement of distribution networks. The DGs, energy storage devices, and switches are optimally coordinated to maximize load restoration. First, an RMPC-based two-layer MG scheduling model considering the worst-case scenario is established to schedule the structure of the MGs and the output of the DGs simultaneously. Furthermore, the original model is transformed into a single-layer programming problem using strong duality theory. The solution of the RMPC-based model can guarantee the reliability and feasibility of MGs within the predefined uncertainty sets of the DGs. The effectiveness of the proposed method is validated with the modified IEEE 37-node distribution test system.
|Journal||International Journal of Electrical Power & Energy Systems|
|Number of pages||19|
|Publication status||Published - 2020|
- Distribution network
- Renewable distributed generator
- Robust model predictive control