A Novel Method for SoH Prediction of Batteries Based on Stacked LSTM with Quick Charge Data

The transition to non-fossil fuels brings with its basic challenges in battery technologies. Due their efficiency, one of the areas where Li-ion batteries are widely used is electric vehicles (EVs). Range estimation most important needs a battery-powered vehicle (BEV). range BEVs directly depends on capacity and powertrain efficiency. Although electrical performance has significantly improved, it still not possible overcome degradation aging. State charge (SoC) state health (SoH) two measures for battery. With accurate SoC SoH estimates, management system can prevent each cell pack from over-charging or over-discharging, prolongs life entire pack. novel idea this study estimate data collected during charging process. needed moment end information, user plan job that will be with. In order meet need, specially designed deep neural network (stacked LSTM) trained tested using measurements only constant current phase quick test results show method effectively applicable chargers.