Data-Driven Fault Recovery With Software-Defined Smart Transmission Grids

This study presents a Software-Defined Transmission Grid (SDTG) framework, integrated with digital and cyber-physical systems, to enable data-driven control within a smart transmission grid. The framework aims to improve grid reliability and accelerate fault recovery. We utilized the Schur complement network reduction technique in the digital model of SDTG to improve grid control through data-driven strategies, with a specific emphasis on rapid system fault recovery. Mathematical proofs based on classic circuit and algebraic graph theories were presented to confirm the effectiveness of the reduced network. Additionally, we introduced a data-driven optimal control approach and demonstrated that the optimal control input can be obtained from a finite set of past control signals, even in the absence of explicit knowledge regarding fault types or the grid's dynamic response. The proposed methodology enhances grid management and offers a flexible and scalable framework for post-fault operation and grid restoration, serving as a foundational approach in other complex transmission grid management scenarios. The effectiveness of this approach was validated through numerical case studies.