Paper presented to the 3rd Southern African Solar Energy Conference, South Africa, 11-13 May, 2015.
The greatest part of Namibia (except the coastal region) receives a large amount of solar radiation throughout the year, yet very few photovoltaic (PV) systems have been installed. The installation of PV systems is associated with high initial capital cost and, typically, an investor would seek a maximum return on investment at minimum risk. To increase the confidence level in the performance of PV systems it becomes imperative that an accurate prediction of the energy yield of systems, operating in the specific Namibian conditions, be achieved. One of the most reliable criterion for the assessment of the performance of a PV system is its annual energy production. The accurate estimation of the energy yield requires the use of a PV simulation software. A number of modelling software packages are currently available that provide a greater or lesser success in accurately predicting the energy yield of PV systems. The success in modelling the yield accurately depends on the skills, knowledge and understanding of the modeller as well as on the uncertainties of the modelling technique. In this work, PVsyst 6 package has been used for modelling an already operational roof-top grid connected PV system. The uncertainties in the modelled and measured data have been assessed and comparison of modelled and measured energy yield has been performed.