A GRU-Based Model for Real-Time Transient Stability Prediction in Electrical Power Systems

Accurate prediction of transient stability is crucial for ensuring the secure and reliable operation of interconnected power systems. However, detecting imminent instability should be fast enough to allow the system controllers to react and maintain the system’s stability. This paper proposes a GRU-based deep learning model for real-time prediction of the system’s stability state following severe disturbances. The performance of the proposed model is evaluated using the IEEE-39 bus system using the DIgSILENT PowerFactory software. Python programming language is adopted to train and test the model, which can seamlessly be integrated with DIgSILENT PowerFactory to perform time-domain simulations. Simulation results demonstrate the efficiency and applicability of the proposed model for real-time stability predictions in power systems.