A Chance-Constrainted Multi-Stage Planning Method for Rural Microgrids Considering Energy Self-Sustained Ability Under Extreme Off-Grid Scenarios

To gradually enhance the energy self-sustained ability and power supply reliability of rural microgrids under extreme off-grid scenarios, this paper proposes a multi-stage chance-constrained planning method. The Autoformer network is used to deeply extract the temporal characteristics of rural loads to realize long-term load forecasting. Then, the adversarial generative network is employed to expand wind and solar data samples to obtain typical scenarios and extreme scenarios. Based on the generated scenarios, a multi-stage planning model is established, and the chance-constraint is used to guarantee the reliable power supply capability under extreme off-grid scenarios. The effectiveness of the proposed method is validated using real data from a rural area in China.