Study On Multi-Health Indicators Fusion And Remaining Useful Life Prediction For Lithium-ion Batteries

Lithium-ion batteries are widely used in electronic equipment,electric vehicles,and large-scale energy storage systems.During the charge-discharge cycles,lithium-ion batteries continuously degrade,resulting in the decrease in capacitance and the increase in internal resistance.Basically,battery is replaced when its capacitance reduces to a certain extent.The remaining useful life(RUL)prediction provides a decision reference for battery health management.In actual engineering,the output power of a lithium-ion battery is affected by either the capacitance or internal resistance.Traditional methods usually carry out the RUL prediction based on capacitance or internal resistance indicator.How to comprehensive ly predict the RUL of lithium-ion batteries is a key scientific problem that needs to be solved in this field.In order to solve the one-sided problem of status-of-health(SOH)estimation of lithium-ion batteries based on a single indicator,the degradation mechanis ms of battery capacitance and internal resistance are analyzed,and then are correlated with constant current charging time(CCCT)and constant voltage charging time(CVCT)to identify the effective indicators,which are capacitance,internal resistance and CCCT.These three indicators are used for the fus ion of SOH of lithium-ion battery to form a new integrated SOH indicator.In order to improve the rationality of the integration of SOH,Beta distribution is employed to fuse the capacitance,internal resistance and CCCT to establish the comprehensive indicators of SOH.The comparisons of SOH estimation based on an individual indicator,and two comprehensive ind icators,and three comprehensive ind icators are carried out.The simulation results presented,the end-of-life(EOL)of a battery based on the fused SOH indicator is shorter than the EOL based on the single SOH indicator,which reveals that battery RUL prediction based on the comprehensive indicators is closer to the engineering compared to the single indicator.On this basis,an improved polyno mial model is used to fit the degradation process of the battery SOH,and the particle filter algorithm is used to predict the RUL.The lifetime probability density of the lithium-ion battery is given.The effectiveness of the model and method for degrading data using lithium-ion batteries is verified.Finally,the RUL prediction results of individual SOH indicator and comprehensive SOH indicators are analyzed.It is recommended that the EOL higher than 0.8 be used as the EOL standard for the comprehensive SOH indicators.