An Imitation Learning-based Approach for Power System Transient Stability Emergency Control

The grid emergency control techniques are facing great challenge with the increasing complexity and uncertainty of modern power systems. To overcome this issue, this paper proposes an imitation learning-based approach for power systems transient stability emergency control. Proposed method can reasonably and effectively utilize the actual existing expert data to enhance the algorithm convergence and improve the control effect. Experimental results demonstrate that the proposed method significantly improves the efficiency and reliability of grid emergency control as well as reduces the manual intervention costs, thus showing great potential for practical application, providing valuable reference in maintaining security and stability of power systems.