PREDICTION OF THE DAILY SOLAR ENERGY, BASED ON THE PATTERN RECOGNITION OF THE RADIATION TIME SERIES, BY THE UTILIZATION OF THE WAVELET TRANSFORMATION AND ARTIFICIAL NEURAL NETWORKS
DOI:
https://doi.org/10.59627/cbens.2010.1702Keywords:
Solar Energy Prediction, Artificial Neural Network, Discrete Wavelet TransformationAbstract
The utilization of the solar energy prediction with low uncertainties is particularly important for the development of intelligent solar heating systems, which are integrated in a demand management system with the objective to suppress to a great extent the peak power of the electric power system. This power demanded by electric showers heads can be reduced with an intelligent preheating strategy of the heating storage, which is accomplished within the early morning hours. As this technologic proposal presents multiples functionalities and economic benefits, the utilization of solar heating systems in residences may become economically feasible in large-scale. The noticeable economic benefit consists in the power reduction of the electric power system within the hours when the main power consumption appears due to the electric shower heads. In the present work is developed a new solar radiation prediction method. It predicts daily solar energy irradiated on horizontal surface, with a prediction horizon of 24 h. The method considers that the behavior of the sky cloud cover of previous days deliver important information for the daily solar radiation prediction. The wavelet transformation is applied to improve the pattern recognition of the time series behavior, separating information of the solar energy and its frequency contents within different frequency bands and day times. With the utilization of the stepwise Multiple Linear Regression (MLR), those model input variables are selected which own statistical significance for the predictions. Furthermore the improvement of the prediction performance is verified by the utilization of an Artificial Neural Network (ANN).
Downloads
References
Addison P.S, 2002. The illustrated wavelet transform handbook - Introductory theory and applications in science, engineering, medicine, and finance, Institute of physics publishing Bristol and Philadelphia, 362 p.
Cao S., Cao J., 2004. Forecast of solar irradiance using recurrent neural networks combined with wavelet analysis, Journal of Applied Thermal Engineering pp. 161-172,
Donaldson K.G., Kamstra M., 1999. Neural network forecast combining with interaction effects, Journal of the Franklin Institute - Pergamon,. 10p.
Daubechies, I., 1992., Ten lectures on wavelets. In: CBMS – NSF Regional Conferences Series in Applied Mathematics SIAM, Philadelphia, 1992, 343 p.
Girodo M., 2006. Solarstrahlungsvorhersage auf der Basis numerischer Wettermodelle, Ph. D Thesis, University of Oldenburg Germany, Faculty of Mathematics and Natural Sciences, Energy and Semiconductor Research Laboratory, 2006, 159 p.
Guarnieri, R. A., 2006. Emprego de Redes Neurais Artificiais e Regressão Linear Múltipla no Refinamento das Previsões de Radiação Solar do Modelo Eta, Dissertação de Mestrado, CPTEC, INPE, São José dos Campos (SP).
Haykin S., 1999. Neural networks – A comprehensive foundation, Prentice Hall, Second Edition, Mc. Master University, Hamilton, Ontario – Canada, , 823 p.
Hocaoglu F.O, Gerek. Ö. N., Kurban M. 2008. Hourly solar radiation forecasting using optimal coefficient 2-D linear filters and feed-forward neural networks, Solar Energy 82, pp.714–726
Kalogirou S. A. 2001. Artificial neural networks in renewable energy systems applications: a review, Renewable and Sustainable Energy Reviews 5 , pp. 373–401
Lorenz, E. 2004. Methoden zur Beschr eibung der Wolkenentwicklung in Satellitenbildern und ihre Anwendung zur Solarstrahlungsvorhersage, Ph.D. thesis, Carl von Ossietzky University, Faculty of Mathematics and Natural Sciences: Oldenburg, Germany, 111 p.
Foresee, F.D., Hagan, M.T., 1997 Gauss-Newton approximation to Bayesian regularization, Proceedings of International Joint Conference on Neural Networks, Nagoya, Japan
Mellit A., Kalogirou S.A. 2008. Artificial intelligence techniques for photovoltaic applications: A review, J., Progress in Energy and Combustion Science, 59 p.
Mellit A., Pavan A.M. 2010. A 24-h forecast of solar irradiance using artificial neural network: Application for performance prediction of a grid-connected PV plant at Trieste, Italy, Solar Energy 84, pp. 807–821
Moreira J.G.S., 1985. Energia solar no Brasil – O uso de coletores planos para o aquecimento de água, Rio de Janeiro, UFRJ, Dissertação de mestrado
Nanavati S.P. e Panigrahi P.K., 2004. Wavelet transform: A new mathematical microscope Editora Springer India, Journal Resonance, Volume 9, Nr. 3 /,15p. http://www.springerlink.com/content/3937325u2mt1kk50/
Oliva G.A., Borges T.P.F., Martins G., Desenvolvimento de tecnologia de pré-aquecedor solar de água para chuveiros elétrica reduzida. Anais do XII Seminário Nacional da Distribuição de Energia Elétrica. Recife, 16 a 21 de Outubro de 1994.
Kratzenberg M.G. 2008. Correção estatística a base da transformada wavelet para previsão de energia solar através de modelo numérico meteorológico, Exame de qualificação, Universidade federal de Santa Catarina
MacKay DJC. A practical Bayesian framework for backpropagation networks. Neural Comput. 1992; 4: 448–472.
Özesmi S.L., Tan C.O., Özesmi U., 2006. Methodological issues in building, training, and testing artificial neural networks in ecological applications, Ecological Modelling 1 9 5 , pp. 83–93
Plump A.P., Rowe R.C., York P., Brown M., 2005. Optimisation of the predictive ability of artificial neural network (ANN) models: A comparison of three ANN programs and four classes of training algorithm, European Journal of Pharmaceutical Sciences 25, pp. 395–405
Salazar J. P.L.C., 2004. Economia de energia e redução do pico da curva de demanda para consumidores de baixa renda por agregação de energia solar térmica, Dissertação, Universidade Federal de Santa Catarina, Departamento de Engenharia Mecânica,
Von Bremen, L., Lisbôa, L. R., Haas, R.; Araujo, F. F. V.; Maciel, C.; Abreu, S. L.; Colle, S., 2006. Local short-term forecasting for wind power plants in Brazil. In: 8th German Wind Energy Conference - DEWEK, Bremen.
World Meteorological Organization (2004), World Climate Research Programme - Baseline Surface Radiation Network - Operations Manual, Version 2.1, 188p.
