To address the nonlinear and non-stationary characteristics of capacity during the aging and degradation process of lithium-ion batteries, a battery capacity prediction model based on particle swarm optimization long short-term memory neural network(PSO-LSTM) was proposed, incorporating changes in charge and discharge characteristics. The key links in the charging and discharging process of lithium-ion batteries were described in detail, including constant current charging time, constant voltage charging time and voltage changes during the discharging process, and the correlation between these characteristics and aging was clarified. Through data mining technology, the variation patterns of these features throughout the battery life cycle were systematically extracted, thereby providing a reliable basis for battery performance evaluation. To verify the effectiveness of the proposed method, multiple sets of experiments were conducted. By comparing with the traditional prediction model, the experimental results show that the maximum root mean squared error (RMSE) of this model in capacity estimation is only 2.417 6. The mean absolute error (MAE) is 1.984 3, demonstrating good adaptability and robustness. It can effectively reflect the actual performance degradation of the battery and provide important theoretical support for the optimization of the battery management system.