Artificial neural network-based method for transient response prediction

The paper describes the application of an ANN-based approach to the prediction of the dynamic behaviour of a synchronous generator following a disturbance in a simple power system. Case data representing the oscillation in rotor angle caused by the disturbance is accumulated from offline simulation using an accurate digital model. Neural networks are trained to map this case data in relation to the initial operating conditions and details of the particular disturbance involved. The response of a one-machine, infinite-bus system is considered after the occurrence of a three-phase short circuit in one element of a double transmission line connecting the synchronous generator to the bus, and of a shock local load change. Numerical results comparing the predicted response using the ANN model with that obtained from direct use of the benchmark model are presented in terms of both accuracy and speed. These suggest that the ANN model might be used in conjunction with an online fault identification system in the study of transient stability or in the provision of information for predictive control. Because only very simple mathematical calculations are required once the neural networks have been trained, the computation time is very short in comparison with the direct use of numerical simulation