Abstract:
Solar energy providers are experiencing increased opportunity in their market as solar and
other renewable energy technologies become more a ordable, accessible and socially encour-
aged. While this means there is opportunity for renewable energy providers, it also lowers the
barriers of entry to the market - creating more competition for existing renewable energy com-
panies. This places an emphasis on e cient project turnover time and low cost power supply to
customers. To achieve both of these goals, a solar energy provider must perform system sizing
quickly (before the customer turns to another provider) and e ectively to ensure they o er the
customer the lowest price possible, while still making su cient return on their investment in a
solar system.
Systems are often conservatively sized by the energy provider to minimise risk of their
minimum return on investment not being met. The study contained herein investigates the
hypothesis of oversizing a solar energy system in the South African market to obtain additional
revenue that outweighs the costs required to oversize a system. A computer model was de-
signed and developed for SolarAfrica to determine the optimal size system for a given site and
consequently, to answer the research hypothesis which was proven correct. Multiple ways of
de ning the optimal system are explored in this report through the maximisation of di erential
income, pro t and customer savings, respectively. Three heuristic-based optimisation methods
(Genetic Algorithm, Particle Swarm Optimisation and Iterative Method) are compared with
regards to their timeliness and e ectiveness in determining an optimal solution.
The report is concluded by selecting the most appropriate objective function and optimi-
sation method for SolarAfrica's case. The ways in which the developed model is anticipated to
create value for SolarAfrica when implemented are also detailed, along with a nal recommen-
dation for future research and testing.
