Transient stability prediction algorithm based on post-fault recovery voltage measurements

This paper presents a novel technique for predicting transient stability status of a power system following a large disturbance. The prediction is based on the synchronously measured samples of the fundamental frequency voltage magnitudes at each generation station. The voltage samples taken immediately after clearing the faults are input to a support vector machine classifier to identify the transient stability condition. The classifier is trained using examples of the post-fault recovery voltage measurements (inputs) and the corresponding stability status (output) determined using a power angle-based stability index. Studies with the New England 39-bus system indicate that the proposed algorithm can correctly recognize when the power system is approaching to the transient instability.