Li-ion battery charging strategy based on multi-state joint estimation model

This paper proposes a fast-charging strategy for lithium-ion batteries based on an intelligent optimization algorithm with multi-physics constraints. First, a thermoelectric coupling model and a battery degradation model have been established. Then an improved particle swarm optimization (PSO) is used to calculate the optimal charging strategy under different weighting factors. Finally, the Pareto boundary is drawn according to the optimization objective, and two charging strategies with different optimization objectives are developed. By comparing in different benchmark functions, the So-improved PSO proposed in this paper improves the convergence of the algorithm compared to the traditional PSO algorithm without increasing the computation time. The simulation results show that the multistage constant-current (MCC) charging strategy proposed in this paper can adjust the magnitude and order of charging current according to the change of weights, and reducing the loss of capacity and energy by only 0.017 % and 35.1 % while improving the charging speed by 24.9 % compared with the constant current-constant voltage (CCCV) charging strategy of 2C. Compared with the traditional CCCV charging strategy, the charging method proposed in this paper can have a broad application prospect. •Research on charging strategy based on a multi-state estimation model•Improved particle swarm algorithm based on improved computational efficiency•Multi-objective optimization considering charging time and energy loss