Hourly global and diffuse radiation of Lagos, Nigeria—Correlation with some atmospheric parameters

• Published in: Solar energy Vol. 52; no. 3; pp. 247 - 251
• Main Authors: Chendo, M.A.C., Maduekwe, A.A.L.
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.03.1994; Elsevier
• Subjects: 140100 - Solar Energy- Resources & Availability; 990200 - Mathematics & Computers; AMBIENT TEMPERATURE; Applied sciences; climatic parameters; CORRELATIONS; DIFFUSE SOLAR RADIATION; EARTH ATMOSPHERE; ENERGIA SOLAR; ENERGIE SOLAIRE; Energy; Exact sciences and technology; FORECASTING; GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; HUMEDAD RELATIVA; HUMIDITE RELATIVE; HUMIDITY; MATHEMATICAL MODELS; MODELE; MODELOS; MULTI-PARAMETER ANALYSIS; Natural energy; NIGERIA; RADIATIONS; RELATIVE HUMIDITY; seasonal analysis; seasonal correlations; solar elevation; SOLAR ENERGY; SOLAR FLUX; SOLAR RADIATION; standard errors; STELLAR RADIATION; VARIATIONS; Nigeria; Africa; Case study; Correlation; Global radiation; Diffuse solar radiation; Solar radiation
• ISSN: 0038-092X; 1471-1257
• DOI: 10.1016/0038-092X(94)90491-X

The influence of four climatic parameters on the hourly diffuse fraction in Lagos, Nigeria, has been studied. Using data for two years, new correlations were established. The standard error of the Liu and Jordan-type equation was reduced by 12.83% when solar elevation, ambient temperature, and relative humidity were used together as predictor variables for the entire data set. Ambient temperature and relative humidity proved to be very important variables for predicting the diffuse fraction of the solar radiation passing through the humid atmosphere of the coastal and tropic city of Lagos. Seasonal analysis carried out with the data showed improvements on the standard errors for the new seasonal correlations. In the case of the dry season, the improvement was 18.37%, while for the wet season, this was 12.37%. Comparison with existing correlations showed that the performance of the one parameter model (namely K t ), of Orgill and Hollands and Reindl, Beckman, and Duffie were very different from the Liu and Jordan-type model obtained for Lagos.