Comparison of methods for the perturbed trajectory prediction based on wide area measurements

With the wide use of phasor measurement unit (PMU) and Wide Area Measurement System (WAMS), the transient instability real time prediction has been one of hot issues in power systems. The perturbed rotor angle trajectory prediction is a key step for transient instability prediction. The curve fitting based perturbed trajectory extrapolation methods had been paid widely attention since they do not need the knowledge of the network topology, model and parameters of power system and do not need network equivalence. Only a few post-fault measurements are required. The existed methods, such as the polynomial model, the trigonometric function model, the auto-regressive model and the angular velocity prediction & integration based method, are overviewed in this paper respectively. The experimental simulations on both IEEE 39-bus system and China Southern Power Grid are done to test the effectiveness of these models. An analysis and comparison is presented in terms of the fault condition, the sampling period, the number of the required measurements and the forecasting accuracy. The numerical results show that the polynomial model based prediction method has best effects for perturbed rotor angle trajectory prediction in terms of stability, rapidity and accuracy.