Generation adequacy assessment of power systems by time series and fuzzy neural network

In this paper, a new hybrid method for generation adequacy assessment of power systems is proposed. This method is based on the combination of time series and fuzzy neural network (FNN). The time series, as a spectrum analyzer, is used to determine the effective input features. The generation adequacy index of loss of energy expectation (LOEE) is selected as the output feature. These input and output features construct the training samples of the FNN. After training, it constructs a mapping function between input and output features, used to predict the adequacy index of the next time interval. Obtained results from extensive testing of the whole method on two practical power systems confirm the validity of the developed approach. It is shown that the proposed method can provide more accurate results than the conventional techniques, such as Monte Carlo simulation, time series, and artificial neural networks.