Cantor: A Novel Dynamic Source‐Grid‐Load‐Storage Dispatching Model for Multiple Objectives in a Regional‐Level Power System

The alignment of the real‐time power supply–demand side becomes a crucial issue in the regional‐level power system due to the growing complexity of source‐grid‐load‐storage scenarios. However, traditional dispatching schemes with inflexible time points are hardly adaptive to the unpredictability of renewable energy generation and multiple‐type load demands. To address this issue, we propose Cantor, a novel dynamic source‐gird‐load‐storage dispatching model for multiple objectives in a regional‐level power system. Cantor incorporates two techniques: (1) An identification algorithm of dispatching time points (IDTP) is designed to satisfy the requirement of power supply–demand alignment dynamically. (2) The latest Twin Delayed Deep Deterministic Policy Gradient decision algorithm is improved to calculate optimal dispatching results in different IDTPs. Furthermore, our case studies highlight Cantor's effective adaptation to complex scenarios in a regional‐level power system dispatching, outperforming state‐of‐the‐art models in power supply reliability, environmental protection, and economic costs, as well as model‐solving efficiency. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.