Optimization of charging process for electric micromobility devices with real-time operation

While fast charging offers undeniable benefits in user convenience, it also presents significant challenges, particularly concerning the thermal management of batteries. The e-micromobility devices, such as e-bikes and e-motorcycle, often use cost-efficiency processors and passive cooling due to cost considerations, which makes it difficult for existing optimization algorithms to run reliably on these devices. To tackle these challenges, this study proposed an optimized charging algorithm for those devices. To meet the need for low computational efforts in real-time usage, the proposed algorithm incorporates both offline training and online operation. During offline training, by considering a trade-off between charging time and temperature variations, optimized charging current policy maps are obtained based on the 1D electrothermal model and dynamic programming. When operation online, the processor only needs to interpolate the optimized policy charging map. The obtained optimized policy maps are further validated in real-time experiments with a 43Ah battery. The experiment results show that, compared to the benchmark 1C constant current charging, the charging time is reduced by 8.5 % with the proposed charging strategies, while the charged State of Charge (SoC) and temperature increases remain nearly identical. Robust analysis demonstrates that the algorithm remains stable and effective under different room-temperature and SoC conditions. Moreover, for real-time operation, a minimum execution time of 0.44 milliseconds on an 8-bit microcontroller demonstrates the method's suitability for low-cost e-micromobility application.