Hybrid Model-Driven and Data-Driven Surrogate-Assisted Method for Transient Stability Preventive Control

Fast and effective transient stability analysis (TSA) and preventive control play an important role in the secure and stable operation of power systems. However, the conventional approach, which relies on offline simulations and critical anticipated fault sets, struggles to adapt to the dynamic and intricate nature of power system dynamics. To address this challenge, this paper introduces a novel method for transient stability preventive control (TSPC). Leveraging a hybrid model-driven and data-driven surrogate-assisted model, our proposed method facilitates real-time formulation of efficacious control strategies. First, the proposed method constructs an analytical representation model to encapsulate transient stability constraints, which is coupled with an optimal power flow solution approach that can accommodate multiple stability constraints. Subsequently, an online correction mechanism for transient stability constraints is introduced to enhance the adaptability of TSA in novel scenarios. Finally, the TSPC model, which considers stability constraints, is transformed into a solvable optimization problem. The method's efficacy was validated using the New England 10-machine 39-node system and the IEEE 50-machine 145-node system. The results of the case studies demonstrate promising computational efficiency and practical applicability.