Study Of Aging Distribution And State Estimation Of Lithium-Ion Batteries

Lithium-ion batteries are the main power supply for electric vehicles and hybrid vehicles.The battery nonuniformity will induce the inhomogeous of current and temperauter which can induce the aging distribution in battery.These aging distribution can increase the aging rate of battery.Even worse,these aging disrribution will induce serious safety issues.By comprehending the characteristics and reasons of aging distribution,it is possible to prolong the cycle lifetime and safety performance of battery through better battery design and high-efficiency cycling control.In this paper,the aging behavior of battery unit and battery module is explored by experimental design.Then,the characteristics and reasons of aging distribution in battery unit and battery module are analyzed.Finally,the safety performance and state estimation of aged battery are explored.Firstly,the aging behavior and mechanism of battery cycling at low temperature and large charging current rate are explored,respectively.Using the principal effect analysis method,the main aging factors cycling at low temperature are identified.The aging mechanism of batteries are confirmed using the external characteristic analysis and postmortem experiments.The main aging mechanism of batteries cycling at low temperature is lithium plating.And the main aging mechanism of batteries cycling at large charging current is SEI growth and manganese atom dissolved.Secondly,the characteristics of aging distribution and reasons in battery unit and battery module are explored with postmortem experiments.In battery unit,middle layers and the positions of tap and center are aged more serious.In battery module,the center batteries are aged more serious.By combining the experiments and the simulation results,the main reason of aging distribution is the distribution of current and temperature in battery unit.And the main reason of aging distribution is the distribution of temperature in battery module.Thirdly,the effects of lithium plating morphology,SEI growth and manganese dissolution on battery safety performance are explored by accelerating rate calorimeter(ARC).Both dendrite and mossy of plated lithium can significantly reduce battery safety performance.Both morphology of plated lithium batteries have similar self heating onset temperatures and thermal runaway onset temperature.As for the aging mechanism of SEI growth and manganese dissolution,batteries cycling at different temperature have similar self heating onset temperature and thermal runaway onset temperature.However,batteries cycling at higher temperatures has shorter time from self heating onset to thermal runaway.The evolution of electrochemical impedance spectroscopy(EIS)in the process of temperature increasing before theremal runaway are explored.The relationship of phase angle at specific frequency and temperature can be used as the characteristic parameter in the medium frequency.The determination coefficient of the linear fitting can be used as the characteristic parameter in low frequency impedance.The error of circle fitting can be used as the characteristic parameter in high frequency impedance.All the characteristic parameter in three frequency range can provide forewarning information for battery safety state.Finally,based on the difference of modules in the battery pack,a state of charge(SOC)estimation algorithm which combined the adaptive extended Kalman filter(AEKF)method and Ampere-Hour integral(AHI)method is developed which can quickly accurate estimate every module SOC at the same time.Due to the SOC estimation accuracy is sensitive to the battery capacity retention and the difficulty of capacity retention estimation,the on-line method of estimating capacity retention based on the normalized feature of discharge curve is proposed.The proposed method can improve the SOC estimation accuracy of aged batteries.