Short term daily load forecasting using recursive ANN

The aspects presented in the paper refer to recursive artificial neural network (RANK) architecture for short term daily load forecasting. The paper describes the training set choice used to teach the RANN and offers the learning method used that insures quick load dynamics learning by the ANN. Using specific data from Banat region (situated in southwestern Romania), some daily load forecasts based on the proposed method are presented and analyzed. On this basis, many useful recommendations are outlined.