Post-Fault Transient Stability Assessment with LTD Compressed Power Angle Trajectories Considering Limitation of Transient Response Observations

The post-fault transient stability assessment (TSA) has shown great importance for the security operation of power systems. Due to the limitation of phasor measurement unit installation and internet communication, it is usually hard to obtain the whole observations across the transient process after fault occurs. Meanwhile, there exists much difficulty in directly predicting the power angle trajectories. In this paper, an improved maximal relevance and minimal redundancy (imRMR) algorithm is put forward for selecting the most crucial transient features so as to reduce the dependency on transient observations. Then the largest triangle dynamic (LTD) method is utilized to compress the power angle trajectories for simplicity. Moreover, the power angle compressed trajectory prediction model is built by Bi-GRU with selected features as input, by which TSA is achieved according to the predicted results. Case studies on CEPRI-TAS system test the validity and efficiency of the proposed scheme.