Lithium-ion batteries (LIBs) have been widely used in various fields. In order to ensure the safety of LIBs, it is necessary to accurately estimate of the state of health (SOH) of the LIBs. This paper proposes a SOH hybrid estimation method based on incremental capacity (IC) curve and back-propagation neural network (BPNN). The voltage and current data of the LIB during the constant current (CC) charging process are used to convert into IC curves. Taking into account the incompleteness of the actual charging process, this paper divides the IC curve into multiple voltage segments for SOH prediction. Corresponding BP neural network is established in multiple voltage segments. The experiment divides the LIBs into five groups to carry out the aging experiment under different discharge conditions. Aging experiment data are used to establish the non-linear relationship between the decline of SOH and the change of IC curve by BP neural network. Experimental results show that in all voltage segments, the maximum mean absolute error does not exceed 2%. The SOH estimation method proposed in this research makes it possible to embed the SOH estimation function in battery management system (BMS), and can realize high-precision SOH online estimation.

Accurate state of health (SOH) estimation is critical to the operation, maintenance, and replacement of lithium-ion batteries (LIBs), which have penetrated almost every aspect of our life. This paper introduces a new approach to accurately estimate the SOH for rechargeable lithium-ion batteries based on the corresponding charging process and long short-term memory recurrent neural network (LSTM-RNN). In order to learn the mapping function without employing battery models and filtering techniques, the LSTM-RNN is initially fed into the health indicators (HIs) extracted from the charging process and trained to encode the dependencies of the related data sequence. Subsequently, the trained LSTM-RNN can properly estimate online SOHs of LIBs using extracted HIs. We experiment on two public datasets for model construction, validation, and comparison. Conclusively, the trained LSTM-RNN achieves an overall root mean square error (RMSE) lower than 1% on the cases with the same discharging current rate and an RMSE of 1.1198% above 80% SOH on another testing case that underwent a different discharging current rate.