Multi-stress accelerated aging for cycle life evaluation of high-capacity, long-life Lithium Iron phosphate batteries

The cycle life assessment of long-life, high-capacity lithium iron phosphate batteries is essential for deployment and operation of reliable energy storage systems. However, conventional testing and evaluation methods are often time-consuming. Developing efficient accelerated aging tests with mechanistic consistency, together with predictive models that correlate to normal aging lifespan, is fundamental to addressing the cycle life assessment of long-life, high-capacity batteries. This paper proposes a lifespan prediction method that integrates multi-stress accelerated aging tests with a segmented degradation model. The full-factorial accelerated aging tests was implemented, covering the temperatures from 55 °C to 75 °C and charge-discharge C-rates from 1C to 2C. Incremental capacity analysis confirmed mechanistic consistency across high-temperature and high-rate conditions, with degradation primarily dominated by loss of lithium inventory. A quantitative temperature-degradation rate relationship was established using the Arrhenius equation, while an empirical model was formulated to characterize the C-rate effect. To capture distinct degradation regimes, a segmented degradation model was proposed, where early-stage nonlinear degradation rate was fitted with a power-law model using the data from the first 300 cycles and steady-state linear degradation rate was extrapolated from accelerated testing data. This method enabled prediction of 880 days (equivalent to 3750 cycles) of aging behaviours with only 90 days of testing. Validation results confirmed high prediction accuracy with all the errors below 4 % at the test endpoint (SOH < 87 %). The proposed method significantly reduces testing cycles and provides an efficient solution to rapid lifespan evaluation of long-life, high-capacity batteries for grid-scale storage.