Abstract:
Solar-powered vehicles are anticipated to have a significant impact on
transportation in the near future, owing to developments in solar energy harvesting
technologies. Nevertheless, there are significant obstacles that must be overcome
in order to make solar-powered vehicles a feasible and viable option. One such
problem is the requirement for precise energy generation forecasting. Accurately
forecasting the energy production for each trip is crucial to guarantee the
continuous functioning of a solar-powered vehicle. Regrettably, current energy
generation forecasting techniques are predominantly tailored for fixed solar panels
and are inadequate for predicting energy generation in mobile, solar-powered
vehicles. This paper proposes a method for predicting the energy produced by
solar-powered cars while they are moving, together with a computerized user
interface for delivering pertinent information. The process encompassed the
identification of all variables influencing solar energy generation and the
acquisition of up-to-the-minute data via application programming interfaces
(APIs). This approach has the potential for worldwide use, as it takes into account
route, topography, and weather data. Its accuracy was evaluated across several
routes, days, hours, speeds, and precision levels in a specific area in Sri Lanka,
using local route, terrain, and weather data. The findings demonstrated that
implementing the procedure on a global scale yields a notable level of precision
at a reduced computational timeframe, whereas implementing it locally produces
comparable accuracy but necessitates a relatively longer computational duration.
