IEEE IECON2025 SYPA Winner - Decentralized Data-based Control for Nonlinear Interconnected Microgrid скачать в хорошем качестве

IEEE IECON2025 SYPA Winner - Decentralized Data-based Control for Nonlinear Interconnected Microgrid

This video presentation describes the work in the paper titled Decentralized Data-based Control for Nonlinear Interconnected DC Microgrids Authors Zhihao Song (Shanghai Jiao Tong University), Li Jin (Shanghai Jiao Tong University) Paper Abstract A novel data-based decentralized control algorithm based on value iteration for nonlinear interconnected DC microgrids. The nonlinearity of the system is illustrated through the inclusion of constant power loads. The proposed algorithm addresses unknown nonlinear dynamics and ensures voltage stability by integrating advanced techniques from nonlinear system control. Specifically, it combines three key components: (1) adaptive dynamic programming for nonlinear optimal control, (2) a decentralized framework for interconnected systems, and (3) integral reinforcement learning to handle unknown dynamics. By employing function approximators, the decentralized algorithm reduces computational complexity, enabling rapid deployment. Simulations on a three-sub-system interconnected DC microgrid demonstrate the algorithm’s effectiveness. Using linear function approximation, the approach successfully derives a stabilizing control solution for the microgrid.