Estimativa da irradiação solar global com partição mensal e sazonal para a região de Cascavel - PR
Carregando...
Data
2005-04-01
Autores
Valiati, Melania I. [UNESP]
Ricieri, Reinaldo P.
Título da Revista
ISSN da Revista
Título de Volume
Editor
Associação Brasileira de Engenharia Agrícola (SBEA)
Resumo
No presente trabalho, mostram-se equações de estimativa da irradiação solar global (R G), por meio do modelo de Angstrom, com partições sazonal e mensal para a região de Cascavel - PR. Os dados experimentais foram cedidos pelo IAPAR, coletados na sua estação meteorológica localizada na COODETEC/Cascavel - PR, no período de 1983 a 1998. Dos 16 anos de dados, 12 anos foram utilizados para cálculo dos coeficientes (a e b) e quatro anos para a validação das equações. Os coeficientes de determinação encontrados foram superiores a 80% para as duas partições. O mínimo da R G é superestimado e o máximo é subestimado quando comparados com o mínimo e o máximo para dados reais, sendo esses encontrados no solstício de inverno e equinócio de primavera, respectivamente. A variação sazonal e mensal do coeficiente a foi menor (0,16 a 0,19 e 0,14 a 0,21) e do coeficiente b maior (0,34 a 0,43 e 0,32 a 0,44). As maiores variações dos erros médios diários ocorreram no equinócio de primavera (-19,45% a 27,28%) e as menores no equinócio de outono (-11,32% a 10,61%). O ajuste mais eficaz das equações foi encontrado para a partição mensal.
This research shows equations of evaluation of global solar irradiation (GR) through Angstrom's model with seasonal monthly partitions for the region of Cascavel -PR, Brazil. The experimental data were given by IAPAR and collected in its meteorological station located in COODETEC/Cascavel - PR, Brazil, from 1983 to 1998. Twelve of the sixteen years of data were used for calculation of the coefficients (a and b) and four years for the validation of the equations. The determination coefficients found were superior to 80% for both partitions. The minimum (GR) is overestimated and the maximum is underestimated when compared to the minimum and maximum for real data, been these found in the winter solstice and spring equinox, respectively. The seasonal and monthly variation of the coefficient a was lower (0.16 to 0.19 and 0.14 to 0.21) and the variation for the coefficient b was higher (0.34 to 0.43 and 0.32 to 0.44). The highest variations of daily average errors had occurred in the spring equinox (-19.45% to 27.28%) and the lowest ones in the autumn equinox (-11.32% to 10.61%). The most efficient adjustment of the equation was found for the monthly partition.
This research shows equations of evaluation of global solar irradiation (GR) through Angstrom's model with seasonal monthly partitions for the region of Cascavel -PR, Brazil. The experimental data were given by IAPAR and collected in its meteorological station located in COODETEC/Cascavel - PR, Brazil, from 1983 to 1998. Twelve of the sixteen years of data were used for calculation of the coefficients (a and b) and four years for the validation of the equations. The determination coefficients found were superior to 80% for both partitions. The minimum (GR) is overestimated and the maximum is underestimated when compared to the minimum and maximum for real data, been these found in the winter solstice and spring equinox, respectively. The seasonal and monthly variation of the coefficient a was lower (0.16 to 0.19 and 0.14 to 0.21) and the variation for the coefficient b was higher (0.34 to 0.43 and 0.32 to 0.44). The highest variations of daily average errors had occurred in the spring equinox (-19.45% to 27.28%) and the lowest ones in the autumn equinox (-11.32% to 10.61%). The most efficient adjustment of the equation was found for the monthly partition.
Descrição
Palavras-chave
heliógrafo, actinógrafo, modelo de Angstrom, actinographer, heliographer, model of Angstrom
Como citar
Engenharia Agrícola. Associação Brasileira de Engenharia Agrícola, v. 25, n. 1, p. 76-85, 2005.