New AI Framework Optimizes Electric Taxi Charging Under Solar Power Uncertainty
As cities increasingly adopt electric vehicles (EVs) for public transportation, a new study proposes a smart solution to one of the sector’s biggest challenges: managing unpredictable solar energy while ensuring efficient service. Researchers have developed a data-driven coordination framework—called SAC—that intelligently dispatches electric taxis for passenger service or charging, depending on real-time solar generation forecasts and passenger demand.
The study, “Stochastic Model Predictive Control-based Electric Taxi Fleet Coordination under Solar Power Uncertainty,” addresses how to optimize the use of solar-powered charging stations and reduce strain on the electric grid. By integrating renewable energy forecasting into its decision-making process, SAC can co-optimize taxi fleet operations and energy usage, tackling issues like solar under-utilization and power grid reliability.
The researchers extend their framework to include stochastic model predictive control (MPC), making it adaptive to the inherent uncertainty in solar energy generation. Evaluations show that SAC boosts solar power utilization by up to 172.6% while maintaining high service quality, with only a 2.2% dip in the supply-demand ratio.
This innovation marks a significant step toward sustainable and scalable electric transportation infrastructure.