Machine learning techniques have been widely used in transient stability prediction of power systems. When using the post-fault dynamic responses, it is difficult to draw a definite conclusion about how long the duration of response data used should be in order to balance the accuracy and speed. Besides, previous studies have the problem of lacking consideration for the confidence level. To solve these problems, a hierarchical method for transient stability prediction based on the confidence of ensemble classifier using multiple support vector machines (SVMs) is proposed. Firstly, multiple datasets are generated by bootstrap sampling, then features are randomly picked up to compress the datasets. Secondly, the confidence indices are defined and multiple SVMs are built based on these generated datasets. By synthesizing the probabilistic outputs of multiple SVMs, the prediction results and confidence of the ensemble classifier will be obtained. Finally, different ensemble classifiers with different response times are built to construct different layers of the proposed hierarchical scheme. The simulation results show that the proposed hierarchical method can balance the accuracy and rapidity of the transient stability prediction. Moreover, the hierarchical method can reduce the misjudgments of unstable instances and cooperate with the time domain simulation to insure the security and stability of power systems.

To achieve rapid real-time transient stability prediction, a power system transient stability prediction method based on the extraction of the post-fault trajectory cluster features of generators is proposed. This approach is conducted using data-mining techniques and support vector machine (SVM) models. First, the post-fault rotor angles and generator terminal voltage magnitudes are considered as the input vectors. Second, we construct a high-confidence dataset by extracting the 27 trajectory cluster features obtained from the chosen databases. Then, by applying a filter–wrapper algorithm for feature selection, we obtain the final feature set composed of the eight most relevant features for transient stability prediction, called the global trajectory clusters feature subset (GTCFS), which are validated by receiver operating characteristic (ROC) analysis. Comprehensive simulations are conducted on a New England 39-bus system under various operating conditions, load levels and topologies, and the transient stability predicting capability of the SVM model based on the GTCFS is extensively tested. The experimental results show that the selected GTCFS features improve the prediction accuracy with high computational efficiency. The proposed method has distinct advantages for transient stability prediction when faced with incomplete Wide Area Measurement System (WAMS) information, unknown operating conditions and unknown topologies and significantly improves the robustness of the transient stability prediction system.

A machine learning-based approach is proposed to predict the transient stability of power systems after a large disturbance. The post-disturbance trajectories of generator rotor angles are taken as a whole cluster, and 19 cluster features are defined to depict the overall transient stability characteristics of the power systems. A hybrid approach, which combines the linear support vector machine with the decision tree, is proposed to generate the final transient stability classifier. Comprehensive studies are conducted on the IEEE 39-bus and IEEE 145-bus test systems to verify the performance of the proposed approach. Test results show that by using the cluster features and the proposed approach, the transient stability of the power system can be predicted accurately with a shorter training time. Furthermore, the prediction classifier is robust to unknown load levels and network topologies, especially under situations when some generator measurements are unavailable and the number of input cluster features is independent of the system scale, making the proposed approach more suitable to transient stability prediction of large-scale power systems.