Transient stability multi-swing step-out prediction with online anomaly detection

Electric power systems are becoming more complex making them more difficult to control. Owing to recent developments in information and communication technologies, power system data have become available online. In this paper, we propose a method that can predict transient stability multi-swing step-out using 'anomaly detection with data mining'. In particular, we focus our attention on the theory of ChangeFinder, which uses the SDAR algorithm and the two-stage learning model. The generator phase angles are obtained from transient stability simulations. They are passed as inputs to the ChangeFinder and the multi swing step-out can be detected. The validity of the proposed method is verified through simulations on the IEEJ 10 machine 47 bus-system.