Energy optimisation of a solar vehicle for South African conditions
Loading...
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Batteries typically used in electric vehicles have up to one hundred times lower energy-to-weight ratios (net calorific value, Wh/kg) than most fossil fuels used in internal combustion engines. Furthermore, solar-powered electric vehicles rely mostly on harvesting energy from solar irradiation (emitted by the sun) which is weather dependant. The need to optimise the energy usage of battery assisted solar-powered electric vehicles is crucial to maximise the range of such a vehicle. If the distance is fixed, weather conditions constant, and the topography flat, one can approximately optimise the energy usage of a solar car by finding the speed, which on average optimises energy usage over an entire journey. Such an approach provides a good guestimate for example for the Bridgestone World Solar Challenge (BWSC) in Australia (taking place primarily in the desolate outback regions) where the topography is predominantly flat, and the weather conditions are relatively constant and therefore very predictable. Contrastingly, the Sasol Solar Challenge (SSC) route in South Africa contains complex topography, which includes various mountainous regions with steep slopes and frequent changes in the gradient of the road. Also, events such as the BWSC, are fixed distance routes, where the SSC is a variable distance route.Various methods to optimise the energy usage of a solar car in a BWSC setting have been proposed. However, none of these methods are not suitable for the SSC context. This work is devoted to both theoretical research and novel applications of bi-level optimisation techniques to minimise energy usage and maximise distance travelled by a solar car participating in an SSC event in South Africa.The early chapters of the work create a foundation in terms of the current state-of-the-art when considering vehicle modelling and interpretation of weather forecasts. A detailed solar car energy model is devised, and Model Output Statistics (MOS) are employed to i