Input space to neural network based load forecasters

The literature has presented a great variety of methodologies for electricity load forecasting. This fact reflects the possible consequences of electricity load forecasting errors, which can cause blackouts with serious social and economic implications. In recent years, papers on this topic have shown the potential of neural network based models. For short term horizons in particular, this capability is explained by the neural networks' flexibility in capturing nonlinear interdependencies between the load and exogenous variables. However, one issue in the development of neural network based electricity load forecasts has not been treated with proper care. Input space representation and complexity control interact strongly, and new solutions for this problem are needed. This paper proposes original automatic procedures for selecting input variables in feedforward neural network based load forecasting models. Filter (model independent) and wrapper (model dependent) approaches have both been investigated in the context of short term load forecasting. Regarding the filter approach, a procedure for identifying relevant subspaces is introduced. For each relevant subspace, interdependences based on a normalized measure of mutual information are estimated to select input variables. Furthermore, phase-space embedding, based on Takens' theorem, is applied in combination with multi-resolution decomposition via wavelet transformation. The results have been compared with a Bayesian wrapper, in which probes have been employed as input references of usefulness.

[1]  Ming-Wei Chang,et al.  Load Forecasting Using Support Vector Machines: A Study on EUNITE Competition 2001 , 2004, IEEE Transactions on Power Systems.

[2]  A.P. Alves da Silva,et al.  Toward Estimating Autonomous Neural Network-Based Electric Load Forecasters , 2007, IEEE Transactions on Power Systems.

[3]  F. Takens.  Detecting strange attractors in turbulence , 1981 .

[4]  Roberto Battiti,et al.  Using mutual information for selecting features in supervised neural net learning , 1994, IEEE Trans. Neural Networks.

[5]  José Ramón Cancelo,et al.  Forecasting the electricity load from one day to one week ahead for the Spanish system operator , 2008 .

[6]  Robert J. Marks,et al.  An adaptively trained neural network , 1991, IEEE Trans. Neural Networks.

[7]  Grantham K. H. Pang,et al.  A probabilistic associative memory and its application to signal processing in electrical power systems , 1992 .

[8]  Fraser,et al.  Independent coordinates for strange attractors from mutual information. , 1986, Physical review. A, General physics.

[9]  H. Kantz,et al.  Nonlinear time series analysis , 1997 .

[10]  Felix F. Wu,et al.  Applied Mathematics for Restructured Electric Power Systems , 2005, IEEE Transactions on Automatic Control.

[11]  T. Senjyu,et al.  Several-hours-ahead electricity price and load forecasting using neural networks , 2005, IEEE Power Engineering Society General Meeting.

[12]  P. McSharry,et al.  Short-Term Load Forecasting Methods: An Evaluation Based on European Data , 2007, IEEE Transactions on Power Systems.

[13]  S. J. Kiartzis,et al.  Short term load forecasting using fuzzy neural networks , 1995 .

[14]  Agnaldo J. R. Reis,et al.  Feature extraction via multiresolution analysis for short-term load forecasting , 2005, IEEE Transactions on Power Systems.

[15]  D. Mackay,et al.  Bayesian methods for adaptive models , 1992 .

[16]  L. S. Moulin,et al.  Confidence intervals for neural network based short-term load forecasting , 2000 .

[17]  Otávio Augusto S. Carpinteiro,et al.  A hierarchical neural model in short-term load forecasting , 2004, Appl. Soft Comput..

[18]  Saifur Rahman,et al.  Input variable selection for ANN-based short-term load forecasting , 1998 .

[19]  J. Vermaak,et al.  Recurrent neural networks for short-term load forecasting , 1998 .

[20]  Alireza Khotanzad,et al.  ANNSTLF-Artificial Neural Network Short-Term Load Forecaster- generation three , 1998 .

[21]  Andrew K. C. Wong,et al.  An event-covering method for effective probabilistic inference , 1987, Pattern Recognit..

[22]  Carlos E. Pedreira,et al.  Neural networks for short-term load forecasting: a review and evaluation , 2001 .

[23]  R. Ramanathan,et al.  Short-run forecasts of electricity loads and peaks , 1997 .

[24]  Gérard Dreyfus,et al.  Ranking a Random Feature for Variable and Feature Selection , 2003, J. Mach. Learn. Res..

[25]  Te Sun Han,et al.  Multiple Mutual Informations and Multiple Interactions in Frequency Data , 1980, Inf. Control..

[26]  Gerd Pfister,et al.  Optimal Reconstruction of Strange Attractors from Purely Geometrical Arguments , 1990 .