Comparação da Acurácia de Modelos de Redes Neurais Artificiais na Predição da Irradiância Solar e Geração de Energia Fotovoltaica

• Carlos Alejandro Urzagasti UNILA
• Joylan Nunes Maciel UNILA
• Victor Hugo Wentz UNILA
• Jorge Javier Gimenez Ledesma UNILA
• Oswaldo Hideo Ando Junior UNILA

DOI: https://doi.org/10.5753/latinoware.2021.19905

Abstract

One of the ways to supply the growing consumption of electricity with the use of clean and renewable energy sources, such as solar photovoltaics. However, this type of generation has intermittencies that increase the instability and insecurity of the electric system. One of the solutions to this problem is the study of methods for Prediction of Solar Photovoltaic Energy Generation (PGESF). In this context, the present study compared the prediction accuracy of Artificial Neural Network (ANN) models, published in [14], from two distinct databases (datasets) and three different short-term loss horizons. The results suggest that the use of different meteorological variables and the size of the dataset significantly influence (p-value<0.001) the models’s accuracy. Furthermore, the model’s prediction accuracy decreases as the prediction horizon increases.

Keywords: predição de energia solar fotovoltaica, irradiância solar, Redes Neurais Artificiais

References

ANEEL, "Sistema de Informações de Geração da ANEEL - SIGA," Agência Nacional de Energia Elétrica., 2020. https://bit.ly/2IGf4Q0 (accessed Apr. 26, 2020).

MME, "Plano Decenal de Expansão de Energia 2026 (versão para consulta pública)." Brasília, p. 264, 2017.

M. G. Villalva, Energia Solar Fotovoltaica: Conceitos e aplicações, 2ed ed. São Paulo: Érica, 2015.

K. Lappalainen and S. Valkealahti, "Photovoltaic mismatch losses caused by moving clouds," Sol. Energy, vol. 158, no. October, pp. 455-461, 2017, doi: 10.1016/j.solener.2017.10.001.

J. Marcos, O. Storkël, L. Marroyo, M. Garcia, and E. Lorenzo, "Storage requirements for PV power ramp-rate control," Sol. Energy, vol. 99, pp. 28-35, Jan. 2014, doi: 10.1016/j.solener.2013.10.037.

S. Shivashankar, S. Mekhilef, H. Mokhlis, and M. Karimi, "Mitigating methods of power fluctuation of photovoltaic (PV) sources - A review," Renew. Sustain. Energy Rev., vol. 59, pp. 1170-1184, 2016, doi: 10.1016/j.rser.2016.01.059.

J. Antonanzas, N. Osorio, R. Escobar, R. Urraca, F. J. Martinez-de-Pison, and F. Antonanzas-Torres, "Review of photovoltaic power forecasting," Sol. Energy, vol. 136, pp. 78-111, 2016, doi: 10.1016/j.solener.2016.06.069.

P. Li, K. Zhou, and S. Yang, "Photovoltaic Power Forecasting: Models and Methods," in 2nd IEEE Conference on Energy Internet and Energy System Integration, EI2 2018 - Proceedings, Oct. 2018, pp. 1-6, doi: 10.1109/EI2.2018.8582674.

R. Blaga, A. Sabadus, N. Stefu, C. Dughir, M. Paulescu, and V. Badescu, "A current perspective on the accuracy of incoming solar energy forecasting," Prog. Energy Combust. Sci., vol. 70, pp. 119-144, 2019, doi: 10.1016/j.pecs.2018.10.003.

J. N. Maciel, J. Javier Gimenez Ledesma, and O. Hideo Ando Junior, "Forecasting Solar Power Output Generation: A Systematic Review with the Proknow-C," IEEE Lat. Am. Trans., vol. 19, no. 4, pp. 612-624, Apr. 2021, doi: 10.1109/TLA.2021.9448544.

P. A. C. Rocha, J. L. Fernandes, A. B. Modolo, R. J. P. Lima, M. E. V. da Silva, and C. A. D. Bezerra, "Estimation of daily, weekly and monthly global solar radiation using ANNs and a long data set: a case study of Fortaleza, in Brazilian Northeast region," Int. J. Energy Environ. Eng., vol. 10, no. 3, pp. 319-334, 2019, doi: 10.1007/s40095-019-0313-0.

A. Mellit and A. M. Pavan, "A 24-h forecast of solar irradiance using artificial neural network: Application for performance prediction of a grid-connected PV plant at Trieste, Italy," Sol. Energy, vol. 84, no. 5, pp. 807-821, 2010, doi: 10.1016/j.solener.2010.02.006.

V. H. Wentz, J. N. Maciel, M. N. Kapp, J. J. G. Ledesma, O. H. Ando Junior, and O. H. A. Junior, "Comparação de Modelos de Redes Neurais Artificiais para a Predição da Irradiância Solar na Geração de Energia Fotovoltaica," in Anais do I Congresso Brasileiro Interdisciplinar em Ciência e Tecnologia, 2020, p. 8, [Online]. Disponível em: [link].

J. N. Maciel, V. H. Wentz, J. J. G. Ledesma, and O. H. Ando Junior, "Analysis of Artificial Neural Networks for Forecasting Photovoltaic Energy Generation with Solar Irradiance," Brazilian Arch. Biol. Technol., vol. 64, no. spe, 2021, doi: 10.1590/1678-4324-75years-2021210131.

F. UFSC, "Fotovoltaica - Grupo de Pesquisa Estratégica em Energia Solar Fotovoltaica," 2021. https://fotovoltaica.ufsc.br/sistemas/fotov (accessed Sep. 20, 2021).

H. T. C. Pedro, D. P. Larson, and C. F. M. Coimbra, "A comprehensive dataset for the accelerated development and benchmarking of solar forecasting methods," J. Renew. Sustain. Energy, vol. 11, no. 3, p. 036102, May 2019, doi: 10.1063/1.5094494.

S. Russell and P. Norvig, Artificial Intelligence: A Modern Approach, 3rd editio. India: Pearson Education India, 2015.

S. O. Rezende, Sistemas Inteligentes, 1st ed., vol. 2013. São Paulo: Manole, 2003.

C. A. Gueymard, "A review of validation methodologies and statistical performance indicators for modeled solar radiation data: Towards a better bankability of solar projects," Renew. Sustain. Energy Rev., vol. 39, pp. 1024-1034, 2014, doi: 10.1016/j.rser.2014.07.117.

E. Mining, Python Machine Learning: Understand Python Libraries (Keras, NumPy, Scikit-Lear, TensorFlow) for Implementing Machine Learning Models in Order to Build Intelligent Systems. Amazon Digital Services LLC - KDP Print US, 2019.

E. Bisong, "Google Colaboratory," in Building Machine Learning and Deep Learning Models on Google Cloud Platform, Berkeley, CA: Apress, 2019, pp. 59-64.

Jamovi, "The jamovi project (2021)," Computer Software, 2021.

P. R. B. Guimarães, Métodos quantitativos estatísticos, 1.ed. rev. Curitiba, PR: IESDE Brasil S.A., 2012.

P. A. Morettin and W. de O. Bussab, Estatística básica, 9a edição. São Paulo: Saraivauni, 2017.

J. Hsu, Multiple Comparisons: Theory and Methods, 1st ed. Ohio - USA: Taylor & Francis, 1996.