A Novel Neural Network Approach to Dynamic State Estimation of Generators Subjected to Ageing in Complex Power Systems

In this paper, a neural network based technique for estimating dynamic states of generators in highly complex power systems is presented. The proposed method is independent to the mathematical model of the generators and uses a nonlinear autoregressive neural network with exogenous inputs to estimate dynamic states of the generators. The proposed technique has been compared to particle filter and unscented Kalman filter based schemes previously reported in the literature. The simulation results show superiority of the proposed technique over the two other schemes when parameters of the generators alter. Parameter alterations in generators are practically occur due to environment impacts, ageing of the equipment and so on. The proposed scheme is capable of keeping its accuracy and precision even in the presence of unobservable variances in generator parameters.