Broadband solar radiometric measurements in the greater Durban area.

Abstract

This work comprises a radiometric study of Durban‟s solar resource, utilizing data from the Howard College campus of the University of KwaZulu-Natal (UKZN), and the Solar Thermal Applications Research Laboratory (STARlab) at Mangosuthu University of Technology (MUT), located 17 km away. The study has three aims: first to establish a solar radiometric monitoring network for the greater Durban area, comprising the UKZN Howard College and Westville stations, and the STARlab facility at MUT. The UKZN Westville station is under refurbishment and should be operational by the end of 2011. Data from this station are not included in the study. The instrumentation and acquisition software in use at Howard College and STARlab are described. The stations record global horizontal irradiance (GHI), direct normal irradiance (DNI) and diffuse horizontal irradiance (DHI), measured by an unshaded pyranometer, a normal incidence pyrheliometer and a pyranometer shaded with a stationary band respectively. Second, to test a number of existing radiometric models against measured data gathered at the stations. Radiometric models assist in estimating missing components of radiation at stations that do not measure all three components separately, for reasons of cost. The models investigated included Erbs et al. (1982), Orgill and Hollands (1977), Reindl et al. (1990), Boland et al. (2001), and Skartveit and Olseth (1987) and correction models by Drummond et al. (1956), Le Baron et al. (1990), Batlles et al. (1995), and Muneer and Zhang (2000) to correct the shadow band effect. Third, to compare data from the two operational stations and to investigate potential spatial differences in sun strength arising from micro-climate effects in the greater Durban area. This takes the form of a statistical analysis of the differences in radiometric data recorded simultaneously at the UKZN and STARlab stations. The study found that the recorded difference in GHI over one year was 0.72%, which lies within the instrument measurement accuracy. Therefore no measurable radiometric differences due to microclimate could be detected and, for the period in which data were collected, measurements from Howard College could be used to estimate irradiance patterns for MUT, and vice versa.