State Of Health Diagnosis Of Lithium Ion Batteries Based On The Sensible Parameters

Effective approaches of state of health(SOH)estimation and remaining useful life(RUL)prediction are significant for the efficient utilization and management of lithium ion batteries(LIBs).How to estimate the SOH of LIBs based on the sensible parameters and to further establish accurate life predictio n models has become the focus in this field.Based on the study of the degradation mode of LIBs,universal and reliable SOH estimation methods which utilize the sensible parameters as the input are developed in this paper.The degradation mode of NCM523/graphite cells cycled at different temperature has been studied.The full cells with different SOH were disassembled.The properties of the electrodes were investigated by SEM,XPS,XRD,Raman and the changes of the separator were studied by permeability test.As for cells cycled at 25 oC,a thick solid electrolyte interface(SEI)layer and a small amount of transition metal elements were detected on the anodes,and no cathode electrolyte interface(CEI)film was found.The formation and growth of SEI is the dominated degradation mechanism for this situation.As for cells cycled at 5 oC,serious lithium plating and deposition of transition metal elements were found on the anodes,and a small amount of CEI film was detected on the cathodes.The degradation of cells for this situation is mainly caused by the lithium plating.No structural changes were found for both the anode and the cathode materials.Serious pore blocking which blocks the diffusion of the lithium ions was detected on the deteriorated separator.The influences of degradation,temperature and current rate on the over-potential curves were studied and the correlations between the over-potential and these three factors were described using empirical models.It was found that the discharge overpotential and SOH or temperature was exponentially correlated while the overpotential and current rate was linearly correlated.Then a novel health indicator based on the discharge curves was proposed.Based on the health indicator,the SOH of LIBs which were cycled under different regimes was estimated.The estimation results show that there is rough linearity between the capacity retention and the proposed health indicator.This approach only utilizes the operating voltage-time data of the LIBs during the actual operation process as the input,and requires no additional characterization tests,which makes it valuable for practical application s.The effects of degradation,temperature and state of charge on the resistance,and the effects of degradation,temperature and current rates on the voltage relaxation curves of LIBs were investigated.The results show that the resistance increases with the degradation of LIBs and the deceasing temperature,and the state of charge mainly influences the charge – transfer resistance.The recovery characteristics of the voltage relaxation curves are related to the degradation of LIBs.Based on the analysis mention above,a novel health indicator with the relaxation voltage and open circuit voltage as the inputs was derived based on an equivalent circuit model.A voltage relaxation model for rapid prediction of the open circuit voltage was proposed.Then the SOH of LIBs was estimated fastly by combing the voltage relaxation model and the proposed health indicator.Results show that the method is suitable for the SOH estimation of LIBs operated under dynamic modes since the input parameters are the relaxation voltage.The features of the charging/discharging curves of the LIBs with a certain current rate during the dynamic operation and the ones with a corresponding constant current rate were investigated and compared.Results show that the discharge curves with a certain current rate extracted from the dynamic operation overlap with the ones measured with the same constant current rate,when the other conditions of the LIBs were kept constant.Based on this,the complete discharge curves(voltage-capacity)were simplified and the values of capacity difference versus voltage difference(delta capacity – delta voltage)were obtained.Then an empirical model was proposed to simulate the delta capacity – delta voltage curves and estimate the SOH of LIBs.The accuracy of SOH estimation for LIBs operated under 10 oC temperature fluctuates is higher than 96%.Results suggest that the simplification method and the empirical model can be used to estimate the SOH of LIBs operated with dynamic conditions.