Deterministic annealing clustering for ANN-based short-term load forecasting

This paper presents a clustering method for preprocessing input data of short-term load forecasting in power systems. Clustering the input data prior to forecasting with the artificial neural network (ANN) decreases the prediction errors observed. In this paper, an MLP ANN is used to deal with one-step ahead daily maximum load forecasting, and the deterministic annealing (DA) clustering is employed to classify input data into clusters. The DA clustering is based on the principle of maximum entropy in statistical mechanics to evaluate globally optimal classification. The proposed method is successfully applied to real data. A comparison is made between the proposed and the conventional methods in terms of the average and the maximum prediction errors. The effectiveness of the proposed method is demonstrated through comparison of the real load data with short-term forecasted values.