The appropriate temperature distribution is indispensable to lithium-ion battery module, especially during the fast charging of the sudden braking process. Thermal properties of each battery cell are obtained from numerical heat generation model and experimental data, and the deviation of thermophysical performance is analyzed by K-means clustering and hierarchical clustering to select battery cells with similar performance. Thermal performance of lithium-ion cells under different charging rates is investigated in experiments and the effects of different mini-channel designs discussed using numerical simulation, maximum temperature, maximum pressure, and temperature standard deviation are compared by both numerical calculation and experimental validation. Two kinds of cooling plates are selected, considering the uniformity of temperature distribution and energy consumption, respectively. All of these cooling plate designs have the ability to constrain the maximum temperature and temperature standard deviation within 306 K and 1.2 K, respectively. Additionally, this thermal management system does not need too much energy consumption. In experimental validation, deviation of maximum temperature is measured to be within 2.2 K and difference of temperature standard deviation is also within tolerance.