Previsão Sazonal de Vazões para a Usina Hidrelétrica de Boa Esperança-PI utilizando Redes Neurais Artificiais
DOI:
https://doi.org/10.26848/rbgf.v14.5.p2629-2645Palabras clave:
Previsão de vazão, Rede Neural Artificial, Índice Climático, Usina Hidrelétrica de Boa Esperança.Resumen
Estudos que apresentam previsões climatológicas têm crescido em importância, na medida em que o aumento da demanda de água e energia cria uma necessidade de se obter informações sobre o comportamento futuro das fontes naturais. Este artigo apresenta dois modelos de previsão de vazões médias do Rio Parnaíba, na altura da Usina Hidrelétrica de Boa Esperança, para os períodos de outubro a dezembro e de janeiro a abril, épocas que apresentam os maiores valores médios para essa variável hidrológica no local. Como preditores, foram utilizados índices climáticos relacionados a temperatura da superfície do mar, os quais foram escolhidos através da análise de teleconexões já conhecidas da literatura e das correlações calculadas entre as vazões observadas e os valores médios dos índices com diferentes defasagens. Em seguida, foram construídas Redes Neurais Artificias do tipo Perceptron Multicamadas para realizar as regressões para os dois cenários. O Modelo Out-Dez apresentou bons resultados, com coeficientes de Nash-Sutcliffe e Kling-Gupta de 0,64 e 0,71, respectivamente, para a fase de teste. Já o Modelo Jan-Abr, apesar de retornar um coeficiente de Nash-Sutcliffe menor, de 0,56, também apresentou um desempenho satisfatório, com Kling-Gupta de 0,73. Deste modo, os modelos constituem ferramentas que podem ser de grande valia para a operação da usina e para o planejamento do setor hídrico local.
Seasonal Streamflow Forecast for the Hydroelectric Power Plant of Boa Esperança-PI using Artificial Neural Networks
A B S T R A C T
Studies presenting climatological forecasts have grown in importance, as the increase in the demand for water and energy creates a need to obtain information about the future behavior of natural sources. This paper presents two models of forecasting the average streamflow of the Parnaíba River, at the Boa Esperança Hydroelectric Power Plant, for the periods from October to December and from January to April, times that present the highest average values for this hydrological variable on site. As predictors, climatic indices related to sea surface temperature were used, which were chosen through the analysis of teleconnections already known in the literature and the correlations calculated between the observed streamflows and the average values of the indices with different lags. Then, Artificial Neural Networks of Multilayer Perceptrons type were built to perform the regressions for these two scenarios. The Oct-Dec Model showed good results, with Nash-Sutcliffe and Kling-Gupta coefficients of 0.64 and 0.71, respectively, for the test phase. The Jan-Apr Model, in spite of returning a smaller Nash-Sutcliffe coefficient, of 0.56, also presented a satisfactory performance, with Kling-Gupta of 0.73. In this way, the models are tools that can be of great value for the operation of the plant and for the planning of the local water sector.
Keywords: Streamflow Forecast. Artificial Neural Network. Climate Index. Boa Esperança Hydroelectric Power Plant.
Descargas
Citas
Anastasiadis, A.D.; Magoulas, G.D.; Vrahatis, M. N. New globally convergent training scheme based on the resilient propagation algorithm. Neurocomputing, [s. l], v. 64, p. 253-270, jan. 2005. https://doi.org/10.1016/j.neucom.2004.11.016.
Araújo, C.B.C.; Dantas Neto, S. A.; Souza Filho, F. A. Previsão de vazões para o açude Orós/CE a partir de dados hidrometeorológicos utilizando perceptrons. Revista Brasileira de Meteorologia, [S.L.], v. 30, n. 1, p. 37-46, mar. 2015. https://doi.org/10.1590/0102-778620140048.
Araújo, C.B.C.; Souza Filho, F.A.; Araújo Junior, L. M.; Silveira, C. S. Previsão Sazonal de Vazões para a Bacia do Orós (Ceará, Brasil) Utilizando Redes Neurais e a Técnica De Reamostragem dos K-vizinhos. Revista Brasileira de Meteorologia, [S.L.], v. 35, n. 2, p. 197-207, jun. 2020. FapUNIFESP (SciELO). http://dx.doi.org/10.1590/0102-7786351015.
Araújo, J.; Mamede, G.; Lima, B. Hydrological Guidelines for Reservoir Operation to Enhance Water Governance: application to the brazilian semiarid region. Water, [S.L.], v. 10, n. 11, p. 1628, 12 nov. 2018. MDPI AG. http://dx.doi.org/10.3390/w10111628.
Araujo, W.C.; Esquerre, K.O.; Sahin, O. Building a System Dynamics Model to Support Water Management: a case study of the semiarid region in the brazilian northeast. Water, [S.L.], v. 11, n. 12, p. 2513, 28 nov. 2019. MDPI AG. http://dx.doi.org/10.3390/w11122513.
Bagirov, A.M.; Mahmood, A.A Comparative Assessment of Models to Predict Monthly Rainfall in Australia. Water Resources Management, [S.L.], v. 32, n. 5, p. 1777-1794, 17 jan. 2018. Springer Science and Business Media LLC. http://dx.doi.org/10.1007/s11269-018-1903-y.
Bansal, P.; Gupta, S.; Kumar, S.; Sharma, S.; Sharma, S. MLP-LOA: a metaheuristic approach to design an optimal multilayer perceptron. Soft Computing, [S.L.], v. 23, n. 23, p. 12331-12345, 21 jan. 2019. Springer Science and Business Media LLC. http://dx.doi.org/10.1007/s00500-019-03773-2.
Block, P.J.; Souza Filho, F.A.; Sun, L.; Kwon, H. A Streamflow Forecasting Framework Using Multiple Climate And Hydrological Models. Journal Of The American Water Resources Association (Jawra), [s. l], v. 45, n. 4, p. 828-843, ago. 2009. http://dx.doi.org/10.1111/j.1752-1688.2009.00327.x.
Brocca, L.; Massari, C.; Pellarin, T.; Filippucci, P.; Ciabatta, L.; Camici, S.; Kerr, Y.H.; Fernández-Prieto, D. River flow prediction in data scarce regions: soil moisture integrated satellite rainfall products outperform rain gauge observations in west africa. Scientific Reports, [S.L.], v. 10, n. 1, p. 1, 27 jul. 2020. Springer Science and Business Media LLC. http://dx.doi.org/10.1038/s41598-020-69343-x.
Canchala, T.; Alfonso-Morales, W.; Carvajal-Escobar, Y.; Cerón, W. L.; Caicedo-Bravo, E. Monthly Rainfall Anomalies Forecasting for Southwestern Colombia Using Artificial Neural Networks Approaches. Water, [s. l] 12, no. 9: 2628. 2020 https://doi.org/10.3390/w12092628.
Castro, W.; Oblitas, J.; Santa-Cruz, R.; Avila-George, H. Multilayer perceptron architecture optimization using parallel computing techniques. Plos One, [S.L.], v. 12, n. 12, p. 3, 13 dez. 2017. Public Library of Science (PLoS). http://dx.doi.org/10.1371/journal.pone.0189369.
Chao, W.T.; Young, C.C.; Hsu, T.W.; Liu, W.C.; Liu, C.Y. Long-Lead-Time Prediction of Storm Surge Using Artificial Neural Networks and Effective Typhoon Parameters: Revisit and Deeper Insight. Water, [s. l], 12, n. 9: 2394, ago. 2020. https://doi.org/10.3390/w12092394.
Chen, H.; Lu, F.; He, B. Topographic property of backpropagation artificial neural network: from human functional connectivity network to artificial neural network. Neurocomputing, [S.L.], v. 418, p. 200-210, dez. 2020. Elsevier BV. http://dx.doi.org/10.1016/j.neucom.2020.07.103.
Cordão, M.J.S.; Rufino, I.A.A.; Alves, P.B.R.; Barros Filho, M.N.M. Water shortage risk mapping: a gis-mcda approach for a medium-sized city in the brazilian semi-arid region. Urban Water Journal, [S.L.], v. 17, n. 7, p. 642-655, 8 ago. 2020. Informa UK Limited. http://dx.doi.org/10.1080/1573062x.2020.1804596.
Córdoba-Machado, S.; Palomino-Lemus, R.; Gámiz-Fortis, S.R.; Castro-Díez, Y.; Esteban-Parra, M.J. Seasonal streamflow prediction in Colombia using atmospheric and oceanic patterns. Journal Of Hydrology, [S.L.], v. 538, p. 1-12, jul. 2016. Elsevier BV. http://dx.doi.org/10.1016/j.jhydrol.2016.04.003.
Dai, L.; Cheng, T.F.; Lu, M. Summer Monsoon Rainfall Patterns and Predictability over Southeast China. Water Resources Research, [S.L.], v. 56, n. 2, p. 1-55, fev. 2020. American Geophysical Union (AGU). http://dx.doi.org/10.1029/2019wr025515.
De, S.S.; Chattopadhyay, G.; Paul, S.; Chattopadhyay, S.; Haldar, D.K.. Forecasting the winter shower over India through a neurocomputing approach. Journal Of Hydroinformatics, [S.L.], v. 14, n. 2, p. 512-522, 2 ago. 2011. IWA Publishing. http://dx.doi.org/10.2166/hydro.2011.124.
Essenfelder, A.H.; Larosa, F.; Mazzoli, P.; Bagli, S.; Broccoli, D.; Luzzi, V.; Mysiak, J.; Mercogliano, P.; Valle, F.D. Smart Climate Hydropower Tool: a machine-learning seasonal forecasting climate service to support cost-benefit analysis of reservoir management. Atmosphere, [S.L.], v. 11, n. 12, p. 1305, 1 dez. 2020. MDPI AG. http://dx.doi.org/10.3390/atmos11121305.
Gupta, H.V.; Kling, H.; Yilmaz, K.K.; Martinez, G.F. Decomposition of the mean squared error and NSE performance criteria: implications for improving hydrological modelling. Journal Of Hydrology, [S.L.], v. 377, n. 1-2, p. 80-91, out. 2009. Elsevier BV. http://dx.doi.org/10.1016/j.jhydrol.2009.08.003.
Hawthorne, S.; Wang, Q.J.; Schepen, A.; Robertson, D. Effective use of general circulation model outputs for forecasting monthly rainfalls to long lead times. Water Resources Research, [S.L.], v. 49, n. 9, p. 5427-5436, set. 2013. American Geophysical Union (AGU). http://dx.doi.org/10.1002/wrcr.20453.
Horan, R.; Gowri, R.; Wable, P.S.; Baron, H.; Keller, V.D.J.; Garg, K.K.; Mujumdar, P.P.; Houghton-Carr, H.; Rees, G. A Comparative Assessment of Hydrological Models in the Upper Cauvery Catchment. Water, [S.L.], v. 151, n. 13, p. 1-25, 11 jan. 2021. MDPI AG. http://dx.doi.org/10.3390/w13020151.
Instituto Nacional De Pesquisas Espaciais – INPE. Mapa do Reservatório Boa Esperança. Disponível em: http://geopro.crn.inpe.br/RH_reser_PI_boaesperanca.htm.
Kapanova, K.G.; Dimov, I.; Sellier, J.M.. A genetic approach to automatic neural network architecture optimization. Neural Computing And Applications, [S.L.], v. 29, n. 5, p. 1481-1492, 28 jul. 2016. Springer Science and Business Media LLC. http://dx.doi.org/10.1007/s00521-016-2510-6.
Kim, T.; Shin, J.Y.; Kim, H.; Kim, S.; Heo, J.H. The Use of Large-Scale Climate Indices in Monthly Reservoir Inflow Forecasting and Its Application on Time Series and Artificial Intelligence Models. Water, [S.L.], v. 11, n. 2, p. 374, 21 fev. 2019. MDPI AG. http://dx.doi.org/10.3390/w11020374.
Knoben, W.J.M.; Freer, J.E.; Woods, R.A. Technical Note: Inherent benchmark or not? comparing nash-sutcliffe and kling-gupta efficiency scores. Hydrology And Earth System Sciences, [S.L.], v. 23, n. 10, p. 4323-4331, 25 out. 2019. Copernicus GmbH. http://dx.doi.org/10.5194/hess-23-4323-2019.
Krajewski, W.F.; Ghimire, G.R.; Quintero, F. Streamflow Forecasting without Models. Journal Of Hydrometeorology, [S.L.], v. 21, n. 8, p. 1689-1704, 1 ago. 2020. American Meteorological Society. http://dx.doi.org/10.1175/jhm-d-19-0292.1.
Mizukami, N.; Rakovec, O.; Newman, A.J.; Clark, M.P.; Wood, A.W.; Gupta, H.V.; Kumar, R. On the choice of calibration metrics for “high-flow” estimation using hydrologic models. Hydrology And Earth System Sciences, [S.L.], v. 23, n. 6, p. 2601-2614, 17 jun. 2019. Copernicus GmbH. http://dx.doi.org/10.5194/hess-23-2601-2019.
Nascimento, A.T.P.; Cavalcanti, N.H.M.; Castro, B.P.L.; Medeiros, P.H.A. Decentralized water supply by reservoir network reduces power demand for water distribution in a semi-arid basin. Hydrological Sciences Journal, [S.L.], v. 64, n. 1, p. 80-91, 2 jan. 2019. Informa UK Limited. http://dx.doi.org/10.1080/02626667.2019.1566728.
Nash, J.E.; Sutcliffe, J.V.. River flow forecasting through conceptual models part I — A discussion of principles. Journal Of Hydrology, [S.L.], v. 10, n. 3, p. 282-290, abr. 1970. Elsevier BV. http://dx.doi.org/10.1016/0022-1694(70)90255-6.
National Oceanic And Atmospheric Administration – NOAA. Physical Sciences Laboratory. Climate Indices: Monthly Atmospheric and Ocean Time-Series. Disponível em: https://psl.noaa.gov/data/climateindices/list/.
Operador Nacional Do Sistema Elétrico - ONS. Dados Hidrológicos/Vazões. Endereço eletrônico: http://www.ons.org.br/
Park, J.H.; Kug, J.S.; Li, T.; Behera, S.K.. Predicting El Niño Beyond 1-year Lead: effect of the western hemisphere warm pool. Scientific Reports, [S.L.], v. 8, n. 1, p. 1-8, 8 out. 2018. Springer Science and Business Media LLC. http://dx.doi.org/10.1038/s41598-018-33191-7.
Perera, A.; Rathnayake, U. Impact of climate variability on hydropower generation in an un-gauged catchment: erathna run-of-the-river hydropower plant, sri lanka. Applied Water Science, [S.L.], v. 9, n. 3, 27 mar. 2019. Springer Science and Business Media LLC. http://dx.doi.org/10.1007/s13201-019-0925-9.
Pool, S.; Vis, M.; Seibert, J. Evaluating model performance: towards a non-parametric variant of the kling-gupta efficiency. Hydrological Sciences Journal, [S.L.], v. 63, n. 13-14, p. 1941-1953, 26 out. 2018. Informa UK Limited. http://dx.doi.org/10.1080/02626667.2018.1552002.
Pour, S.H.; Wahab, A. K.A.; Shahid, S. Physical-empirical models for prediction of seasonal rainfall extremes of Peninsular Malaysia. Atmospheric Research, [S.L.], v. 233, p. 104720, mar. 2020. Elsevier BV. http://dx.doi.org/10.1016/j.atmosres.2019.104720.
Pulukuri, S.; Keesara, V.R.; Deva, P. Flow Forecasting in a Watershed using Autoregressive Updating Model. Water Resources Management, [S.L.], v. 32, n. 8, p. 2701-2716, 23 mar. 2018. Springer Science and Business Media LLC. http://dx.doi.org/10.1007/s11269-018-1953-1.
Quinn, D.F.; Murphy, C.; Wilby, R.L.; Matthews, T.; Broderick, C.; Golian, S.; Donegan, S.; Harrigan, S. Benchmarking seasonal forecasting skill using river flow persistence in Irish catchments. Hydrological Sciences Journal, [S.L.], v. 66, n. 4, p. 672-688, 12 mar. 2021. Informa UK Limited. http://dx.doi.org/10.1080/02626667.2021.1874612.
Ramchoun, H.; Idrissi, M.A.J.; Ghanou, Y.; Ettaouil, M. Multilayer Perceptron New Method for Selecting the Architecture Based on the Choice of Different Activation Functions. International Journal Of Information Systems In The Service Sector, [S.L.], v. 11, n. 4, p. 21-34, out. 2019. IGI Global. http://dx.doi.org/10.4018/ijisss.2019100102.
Ramu, D.A.; Rao, S.A.; Pillai, P.A.; Pradhan, M.; George, G.; Rao, D. N.; Mahapatra, S.; Pai, D.S.; Rajeevan, M.. Prediction of seasonal summer monsoon rainfall over homogenous regions of India using dynamical prediction system. Journal Of Hydrology, [S.L.], v. 546, p. 103-112, mar. 2017. Elsevier BV. http://dx.doi.org/10.1016/j.jhydrol.2017.01.010.
Santos, L.; Thirel, G.; Perrin, C. Technical note: pitfalls in using log-transformed flows within the kge criterion. Hydrology And Earth System Sciences, [S.L.], v. 22, n. 8, p. 4583-4591, 30 ago. 2018. Copernicus GmbH. http://dx.doi.org/10.5194/hess-22-4583-2018.
Şen, Z.; Altunkaynak, A.; Özger, M. El Niño Southern Oscillation (ENSO) Templates and Streamflow Prediction. Journal Of Hydrologic Engineering, [S.L.], v. 9, n. 5, p. 368-374, set. 2004. American Society of Civil Engineers (ASCE). http://dx.doi.org/10.1061/(asce)1084-0699(2004)9:5(368).
Silva, M.V.M.; Silveira, C.S.; Costa, J.M.F.; Martins, E.S.P.R.; Vasconcelos Júnior, F.C. Projection of Climate Change and Consumptive Demands Projections Impacts on Hydropower Generation in the São Francisco River Basin, Brazil. Water, [S.L.], v. 13, n. 3, p. 332, 29 jan. 2021. MDPI AG. http://dx.doi.org/10.3390/w13030332.
Silva, W.L.; Xavier, L.N.R.; Maceira, M.E.P.; Rotunno, O.C. Climatological and hydrological patterns and verified trends in precipitation and streamflow in the basins of Brazilian hydroelectric plants. Theoretical And Applied Climatology, [S.L.], v. 137, n. 1-2, p. 353-371, 23 ago. 2018. Springer Science and Business Media LLC. http://dx.doi.org/10.1007/s00704-018-2600-8.
Silveira, C.S.; Alexandre, A.M.B.; Souza Filho, F.A.; Vasconcelos Junior, F.C.; Cabral, Samuellson Lopes. Monthly streamflow forecast for National Interconnected System (NIS) using Periodic Auto-regressive Endogenous Models (PAR) and Exogenous (PARX) with climate information. RBRH, [S.L.], v. 22, p. 1-10, 2017. FapUNIFESP (SciELO). http://dx.doi.org/10.1590/2318-0331.011715186.
Souza Filho, F.A.; Lall, U. Seasonal to interannual ensemble streamflow forecasts for Ceara, Brazil: applications of a multivariate, semiparametric algorithm. Water Resources Research, [S.L.], v. 39, n. 11, p. 1-13, nov. 2003. American Geophysical Union (AGU). http://dx.doi.org/10.1029/2002wr001373.
Souza Filho, F.A.; Lall, U. Modelo de Previsão de Vazões Sazonais e Interanuais. RBRH, [s. l], v. 2, n. 9, p. 61-74, jun.2004. http://dx.doi.org/10.21168/rbrh.v9n2.p61-74.
Souza, S.O.; Lima, G.R.R.; Alencar, F.K.M.; Batista, A.C.O.N.; Silva, F.J.A. Assessment of water quality for industrial and irrigation purposes based on ionic indices in the Sítios Novos Reservoir, Ceará, Brazil. Sustainable Water Resources Management, [S.L.], v. 6, n. 5, 21 set. 2020. Springer Science and Business Media LLC. http://dx.doi.org/10.1007/s40899-020-00452-1.
Tuel, A.; Eltahir, E.A.B. Seasonal Precipitation Forecast Over Morocco. Water Resources Research, [S.L.], v. 54, n. 11, p. 9118-9130, nov. 2018. American Geophysical Union (AGU). http://dx.doi.org/10.1029/2018wr022984.
Tsakiri, K.; Marsellos, A.; Kapetanakis, S. Artificial Neural Network and Multiple Linear Regression for Flood Prediction in Mohawk River, New York. Water, [S.L.], v. 10, n. 9, 29 ago. 2018. MDPI AG. http://dx.doi.org/10.3390/w10091158.
Uvo, C.B.; Foster, K.; Olsson, J. The spatio-temporal influence of atmospheric teleconnection patterns on hydrology in Sweden. Journal Of Hydrology: Regional Studies, [S.L.], v. 34, abr. 2021. Elsevier BV. http://dx.doi.org/10.1016/j.ejrh.2021.100782.
Wang, C.; Lee, S.K.; Enfield, D.B. Atlantic Warm Pool acting as a link between Atlantic Multidecadal Oscillation and Atlantic tropical cyclone activity. Geochemistry, Geophysics, Geosystems, [S.L.], v. 9, n. 5, p. 1-17, maio 2008. American Geophysical Union (AGU). http://dx.doi.org/10.1029/2007gc001809.
Zhang, Q.; Li, Z.; Snowling, S.; Siam, A.; El-Dakhakhni, W. Predictive models for wastewater flow forecasting based on time series analysis and artificial neural network. Water Science And Technology, [S.L.], v. 80, n. 2, p. 243-253, 15 jul. 2019. IWA Publishing. http://dx.doi.org/10.2166/wst.2019.263.
Descargas
Publicado
Cómo citar
Número
Sección
Licencia
Derechos de autor 2021 David Lopes de Sousa, Francisco de Assis de Souza Filho, Rafael Brito Aguiar de Araujo, Guilherme Marques Farias, Marco Aurélio Holanda de Castro

Esta obra está bajo una licencia internacional Creative Commons Atribución 4.0.
Material protegido por derechos de autor y plagio. En caso de material protegido por derechos de autor reproducido en el manuscrito, la atribución completa debe ser informada en el texto; un documento de respaldo de la autorización debe enviarse al Consejo Editorial como documento complementario. Es responsabilidad de los autores, no de la revista o de los editores y revisores, informar en el artículo la autoría de los textos, datos, figuras, imágenes y / o mapas publicados anteriormente en otros lugares. Si existe alguna sospecha sobre la originalidad del material, el Comité Editorial puede verificar el manuscrito en busca de plagio. En los casos en que se confirme el plagio, el manuscrito será devuelto sin más revisión y sin la posibilidad de volver a enviarlo. El autoplagio (es decir, el uso de frases idénticas de documentos previamente publicados por el mismo autor) tampoco es aceptable.






