Transient stability real-time prediction for multi-machine power systems by using observation

This paper presents a fast real-time transient stability prediction method for multimachine power systems which does not need the knowledge of network configuration and load characteristics. The least-squares method is used to estimate the parameters of the model and an adaptive technique is used to improve the prediction accuracy. The computational time is extremely fast and is independent of the number of machines and nodes. The prediction is performed on observation data only, and does not require knowledge of system parameters. The effectiveness and feasibility of the method has been tested on a four-machine power system.<<ETX>>