Short-term bus load forecasting of power systems by a new hybrid method

In this paper, a new hybrid method is proposed for short-term bus load forecasting of power systems. The method is composed of the forecast-aided state estimator (FASE) and the multilayer perceptron (MLP) neural network. The FASE forecasts hourly loads of each bus by means of its previous data. Then the inputs and outputs of the FASE are fed to the MLP neural network. In other words, the MLP is trained to extract the mapping function between the inputs and outputs of the FASE (as input features) and real loads as output features. The proposed hybrid method has been examined on a real power system, a part of Iran's power network. The obtained results, discussed comprehensively, show that the hybrid method has better prediction accuracy than the other methods, such as MLP, FASE, and the periodic auto-regression (PAR) model.