Data-Driven Short-Term Voltage Stability Prediction Based on Pre-Fault Operating State

Voltage stability is one of the key factors to ensure the stable operation of power systems. The study of Short-Term Voltage Stability (STVS) is necessary when the complexity of the dynamic response of the power system is increased. In order to reduce the probability of short-term voltage instability and the consequences caused by short-term voltage instability, it is necessary to predict the STVS of power systems before the faults have cause serious consequences. In this paper, a STVS prediction method based on pre-fault operation status and anticipate faults is proposed. A Support Vector Machine (SVM) model is constructed and trained based on the samples generated using numerical simulation software. The effectiveness of the proposed method is presented based on an actual power grid model of China. The prediction accuracy of the SVM model can reach over than 96%, so it has the potential to use practically.