WAMS（wide area measurement system） can monitor real-time trajectories of power system on unified time scale which makes it possible to analyze the real-time transient stability.Based on the gray system theory and self-memory principle of dynamical system

we proposed a method for the disturbed rotor angle trajectory prediction based on gray Verhulst self-memory model.The trend of disturbed rotor angle trajectory during every swing conforms the sigmoid process of gray Verhulst model.For high randomness and nonlinearity of disturbed rotor angle trajectory

the self-memory gray Verhulst model is built based on the traditional gray Verhulst model prediction.Since the prediction not only considers the change trend of disturbed rotor angle trajectory but also uses the historical data

the random fluctuation of disturbed rotor angle trajectory can be characterized well and the prediction performance is improved.Calculation results of New England 10-machine 30-bus system show that the proposed prediction method is effective and can offer technical supports for real-time transient stability analysis.