Lithium-ion batteries are widely utilized in electric transportation because of their high energy density. However, certain e-mobility applications lack active cooling systems and rely on low-cost processors. This poses challenges for optimizing fast charging and implementing complex optimization algorithms, given the limitations of the cooling system and slower processors.
In this study, a fast-charging optimization algorithm is proposed for electric transportation equipped only with a passive cooling system and a cost-effective processor. The proposed algorithm meets the requirement of only a low computational effort when running in real-time. To do so, the algorithm incorporates both offline training and online operation. In offline training, a comprehensive electro-thermal model is presented, encompassing both reversible and irreversible heat generation within the battery. The charging process is then divided into several stages by different start and end SoC. The model is then used to calculate a state grid based on different initial temperatures, which includes the charging temperature and charging time for each stage of charging. Dynamic programming is then used to find the optimal charging current based on the cost of the state grid to form an optimized track map based on the battery status. In real-time operation, the processor simply interpolates the optimized track map based on the battery’s current state to obtain the optimized charging current. The experiments show that the algorithm reduces the charging time by 8.5% as the State of Charge increases from 10% to 80%, while maintaining the same temperature rise compared to 1C charging. Robust analysis indicates that the algorithm maintains good stability and performance across various initial temperatures and charging intervals. For real-time operation, the algorithm takes only 0.44 and 0.073 milliseconds to operate on an 8-bit microprocessor and 32-bit microprocessor, respectively. These results demonstrate the efficiency and stability of the proposed algorithm.