This paper proposes an adaptive multistage constant current-constant voltage (MCCCV) strategy for charging electric vehicles in different situations. First, a high-fidelity thermoelectric-aging coupling model based on a resistor-capacitor pair electrical model, a thermal network model, and a semi-empirical aging model is constructed. Second, an adaptive MCCCV charging strategy involving optimization of the charging current using particle swarm optimization is developed. It can satisfy the preference of users for reducing the charging time or the battery degradation. Finally, three charging strategies based on the Pareto boundary curve of the battery charging time-state of health are developed: a fast-charging strategy for motorway driving, a minimum-aging charging strategy for family use, and a balanced charging strategy for daily use. Additionally, according to the Pareto boundary, the effects of key factors on the optimization of the charging strategy are analyzed and compared. The results show that the balanced charging strategy is 3.60% better than the 0.5C constant current-constant voltage (CCCV) charging strategy recommended by the battery manufacturer with regard to aging loss. Moreover, the charging time is reduced by 37%. Compared with the traditional CCCV charging strategy, the proposed adaptive MCCCV charging strategy has good application prospects with regard to both the charging time and the battery degradation. (C) 2019 Elsevier Ltd. All rights reserved.