A Novel Fast Transient Stability Assessment of Power Systems Using Fault-On Trajectory
Meysam Shahriyari, Hamid Khoshkhoo, Josep M. Guerrero
Abstract
In this article, for fast transient stability status (TSS) prediction, a new scheme is proposed, which does not require any postfault data and predicts TSS using data measured until the end of the fault-on period. In this method, phasor measurement units (PMUs) are located at synchronous generators' (SGs) buses to gather time-synchronized data. Then, these measurements are applied to the proposed procedure to calculate the feature set utilizing fault-on and prefault information and predict TSS by a decision tree classifier. Since the feature set does not include any postfault (i.e., after short-circuit clearance) data, the proposed method has the capability of fast prediction of postdisturbance TSS, and hence, it provides more opportunity for system operators and/or special protection systems to implement timely corrective actions to prevent instability. Also, the results of the time-domain analysis carried out in IEEE14-bus, IEEE39-bus, 16-Machine (68-bus), and IEEE118-bus test systems and comparison of them with previous techniques reveal that although the proposed method requires fewer PMUs, it can forecast TSS accurately under N-1 and N-2 conditions. Besides, these results confirm that, considering the measurement error, when the rotor angle of SGs is estimated using the data measured by WAMS, the proposed method accurately predicts TSS.