An artificial neural network for estimating transient stability limits

In this paper, the nonlinear mapping relation between the transient energy margin and the generator power at different fault clearing time and load levels of the system was established by using the multi-layer feedforward neural network of the perceptron type. Lyapunov´s direct method was used as a fast method to obtain the training set of the artificial neural network (ANN). The transient stability power limits of the generator at different fault clearing time and load levels of the system were estimated very quickly by ANN. The proposed approach was tested on a 4-generator power system, and the results were found to be quite accurate. By comparison with the analytical sensitivity approach, the proposed method avoids the necessity of finding the analytical sensitivity of the transient energy margin to parameter changes, and can quickly estimate transient stability power limits at different fault clearing time and load levels of the system.