Data-driven robust optimization scheduling model for multi-area interconnected power grid

The uncertain distribution of renewable power generation across multi-area interconnected power grids (MIPGs) necessitates coordination of power exchange among subareas for low-carbon oriented sustainable power integration. However, the limited data share among subareas caused by the requirement of information privacy makes difficulties for MIPGs scheduling. Therefore, this study proposes a data-driven robust optimization model and corresponding parallel solving algorithm. Firstly, a distributionally robust schedule model of MIPGs incorporating intra-area and inter-area power generation and transmission constraints is established. The relations of decision-making variables related to different subareas are decoupled to derive the formulations of the economic dispatching model suitable for task decomposition. Then, a data-driven multi-area ambiguity set is developed by integrating discrete scenario probabilities of separate subareas. The conservatism of the set is adaptively adjusted based on the quantity of historical data. Finally, a distributionally parallel (DP) algorithm is employed to effectively facilitate rapid solution by sufficiently utilizing computational resources. Case studies demonstrate the validity of the proposed model and method.