FORECASTING SOLAR RADIATION IN BRAZILIAN CITIES USING A UNIFIED MULTILAYER PERCEPTRON MODEL

FORECASTING SOLAR RADIATION IN BRAZILIAN CITIES USING A UNIFIED MULTILAYER PERCEPTRON MODEL

Conterato, Flavio Santos; Nascimento Filho, Aloísio S.; Saba, Hugo

Full article:

This comprehensive study investigates the utility of a unified Multilayer Perceptron (MLP) for 1-hour solar radiation forecasting in four significant Brazilian cities: Brasília, Salvador, Manaus, and Porto Alegre. The study's diverse geographical locations ensure a comprehensive evaluation of the MLP model's predictive performance under varying climatic conditions. The unified MLP model exhibited successful performance across all cities, showcasing its adaptability and versatility, with an average MAE of 174.59, Pearson correlation above 0.92, and R² above 0.8. These results offer valuable insights for integrating advanced AI techniques into renewable energy applications, contributing to the sustainable development of solar energy systems.

This comprehensive study investigates the utility of a unified Multilayer Perceptron (MLP) for 1-hour solar radiation forecasting in four significant Brazilian cities: Brasília, Salvador, Manaus, and Porto Alegre. The study's diverse geographical locations ensure a comprehensive evaluation of the MLP model's predictive performance under varying climatic conditions. The unified MLP model exhibited successful performance across all cities, showcasing its adaptability and versatility, with an average MAE of 174.59, Pearson correlation above 0.92, and R² above 0.8. These results offer valuable insights for integrating advanced AI techniques into renewable energy applications, contributing to the sustainable development of solar energy systems.

Palavras-chave: - -

DOI: 10.5151/siintec2023-305767

Referências bibliográficas

• [1] "[1] KABIR, Ehsanul et al. Solar energy: Potential and future prospects. Renewable and Sustainable Energy Reviews, v. 82, p. 894-900, 2018. [2] MELLIT, Adel; PAVAN, Alessandro Massi. A 24-h forecast of solar irradiance using artificial neural network: Application for performance prediction of a gridconnected PV plant at Trieste, Italy. Solar energy, v. 84, n. 5, p. 807-821, 2010. [3] Dias, Palmira Maria Acioli; Aylas, Georgynio Yossimar Rosales; Conterato, Flávio Santos; Santos, José Vicente Cardoso; Vieira, Carolina Sacramento; Nascimento, Erick Giovani Sperandio; Moreira, Davidson Martins; ""ANÁLISE DO POTENCIAL DE ENERGIA SOLAR NO ESTADO DA BAHIA USANDO UMA MODELAGEM COMPUTACIONAL"", p. 556-563 . In: Anais do V Simpósio Internacional de Inovação e Tecnologia. São Paulo: Blucher, 2019. ISSN 2357-7592, DOI 10.5151/siintec2019- 70 [4] Bispo, Amanda Santana Lins; Nascimento, Lian Filipe Santana; Gomes, Yuri de Jesus; Conterato, Flávio Santos; ""COMPARAÇÃO DE MODELOS PARA PREVISÃO DA VELOCIDADE DO VENTO ATRAVÉS DE REDES NEURAIS EM LENÇÓIS, BAHIA"", p. 577-584. In: VII International Symposium on Innovation and Technology. São Paulo: Blucher, 2021. ISSN 2357-7592, DOI 10.5151/siintec2021-208825 [5] Bishop, C. M. Pattern recognition and machine learning. Springer, 2006. [6] RIBEIRO, Maria de Nazaré Góes. Aspectos climatológicos de Manaus. Acta amazonica, v. 6, p. 229-233, 1976. [7] FERRARO, Lilian Maria Waquil; HASENACK, Heinrich. Clima. Universidade Federal do Rio Grande do Sul. Instituto de Biociências. Centro de Ecologia. Carvão e meio ambiente. Porto Alegre: A Universidade, 2000. p. 26-44., 2000. [8] MACIEL, Alexandra Albuquerque et al. Projeto bioclimático em Brasília: estudo de caso em edifício de escritórios. 2002. [9] DE CARVALHO, Antonio Pedro Alves; JÚNIOR, Manoel Messias Teixeira. Clima, meio ambiente urbano e a dengue em Salvador. Revista da Arquitetura: cidade e habitação, v. 1, n. 2, 2021. [10] Goodfellow, I., Bengio, Y., & Courville, A. Deep Learning. MIT Press, 2016. [11] Duarte, Lucas Calil Barbosa; Paixão, Moisés Araújo da; Bastos, Luis Felipe da Fé; Conterato, Flávio Santos; ""COMPARATIVO ENTRE REDES NEURAIS, GERANDO PREVISÕES PARA RADIAÇÃO SOLAR GLOBAL E TEMPERATURA DO AR"", p. 569-576 . In: VII International Symposium on Innovation and Technology. São Paulo: Blucher, 2021. ISSN 2357-7592, DOI 10.5151/siintec2021-208830 [12] GÉRON, Aurélien. Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow. "" O'Reilly Media, Inc."", 2022. [13] Willmott, C. J., & Matsuura, K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Climate Research, 30(1), 79-82, 2005. [14] Conterato, Flavio Santos ; Moraes, Caique Augusto Cardoso de ; Pabst, Grhamm
• [2] Caldas, João Felipe de Araújo; Souza, Matheus Carvalho Nascimento de ; Moreira,
• [3] Davidson Martins ; Nascimento, Erick Giovani Sperandio ; ""COMPARISON OF
• [4] SHORT-TERM TEMPERATURE FORECASTING BETWEEN ARTIFICIAL NEURAL
• [5] NETWORK AND MACHINE LEARNING."", p. 535-542 . In: Anais do VI Simpósio
• [6] Internacional de Inovação e Tecnologia. São Paulo: Blucher, 2020. ISSN 2357-7592,
• [7] ISBN: 2357-7592 DOI 10.5151/siintec2020-COMPARISONOFSHORT-TERM
• [8] Batista, G. E., Prati, R. C., & Monard, M. C. A study of k-nearest neighbour as
• [9] an imputation method. HIS, 87(251-260), 48, 2003.
• [10] Zhang, K., et al. Solar Radiation Prediction Based on Multilayer Perceptron
• [11] Neural Network. Energy Procedia, 182, 153-159, 2022."

Como citar:

Conterato, Flavio Santos ; Nascimento Filho, Aloísio S. ; Saba, Hugo ; "FORECASTING SOLAR RADIATION IN BRAZILIAN CITIES USING A UNIFIED MULTILAYER PERCEPTRON MODEL", p-147-153. In: . São Paulo: Blucher, 2023.
ISSN 23577592, DOI 10.5151/siintec2023-305767