(1) Early life prediction using 100 cycles. The most famous one is the RUL single-point prediction method based on the characteristics of discharge capacity curve proposed by Severson et al. This method takes the mean square value of the discharge capacity curve under different aging states of the battery as a feature.
The model can predict the battery cycle life only using the data of the first 100 cycles (Approximately 10% of overall cycle data). Following this, Attia explored closed-loop optimization methods for fast charging protocols, integrating early-stage life cycle predictions (the first 100 cycles) with Bayesian optimization.
This research provides a reliable method for the analysis and evaluation of the charging and discharging characteristics of lithium batteries, which is of great value for improving the safety and efficiency of lithium battery applications.
Mechanism-guided methods predict battery capacity degradation and lifespan by establishing models based on physical mechanisms or using electrochemical analysis techniques to characterize internal electrochemical reactions. Experience-based methods use empirical formulas to fit the capacity degradation trajectory of LIBs.
Ning et al. studied the battery capacity loss at different discharge rates (1–3C) and found that the largest battery internal resistance could be achieved at the 3C discharge rate, and the capacity loss is proportional to the discharge rate.
In the normal environment and high-temperature environment, the charging and discharging time meets the experimental requirements, and the two batteries have good charging and discharging performance in the normal operating temperature range.