Hybrid Model of Cascading Failure in Power Systems using Trajectory Sensitivity-based Classification

A new hybrid model of cascading failure is proposed, which employs a detailed dynamic model to identify and correct the cascade propagation path of QSS model. At every event or failure, trajectory sensitivities for a few seconds following that failure are computed using the detailed dynamic model. Classification based on computed sensitivities reveal the accuracy of QSS model's prediction. If necessary, corrections are made by simulating the detailed dynamic model for a sufficient time. The proposed hybrid model along with a review of trajectory sensitivity is presented. The effectiveness of the proposed model is validated through case studies on IEEE 4-machine and 16-machine systems.