Research On Collaborative Estimation Of SOC And SOH Of Ternary Lithium Battery For Electric Vehicles

As an important means of solving energy and environmental problems,the development of electric vehicles has been an important trend in the development of the automotive industry.At present,the production and sales of electric vehicles and power batteries in China are in the first echelon in the world,and relevant industries and institutions are focusing on the key theories and technologies of power lithium-ion batteries as their current research priorities.The characteristics of lithium-ion batteries in terms of specific power,specific energy and cycle life are very suitable for the demand of electric vehicles in terms of power and economy.In order to maintain the reliability and high efficiency of lithium-ion batteries in electric vehicles and prolong their service life,the battery state,especially the State of Charge(SOC)and State of Health(SOH),should be accurately estimated and monitored.SOH is an important item for battery management.In this paper,model-based estimation of battery SOC and SOH is investigated for ternary Li-ion batteries.Firstly,by analysing the working principle of the battery,battery tests are carried out,a charge/discharge multiplier test and a temperature test are designed to explore the effect of different factors on the available capacity of the battery.Based on the advantages and disadvantages of different battery models,a second-order equivalent circuit is chosen as the basis of this paper’s model,combined with the open circuit voltage-state of charge(OCV-SOC)relationship curve,using a recursive least squares method based on the forgetting factor to identify the battery model parameters,verifying the effectiveness of the method through working conditions,and investigating the effect of temperature on the battery model parameters.Secondly,considering the vulnerability of battery SOC estimation to noise with various unspecified properties,a noise adaptive method based on extended Kalman filtering is used to reduce the uncertainty of noise on SOC estimation.Through algorithm comparison,it is demonstrated that the model-based Adaptive Extened Kalman Filter(AEKF)has better resistance to complex currents as well as high-current operating conditions,and can improve SOC estimation accuracy and reduce fluctuations.Finally,the influence factors of battery SOH are analysed,and a Dual Extended Kalman Filter(DEKF)with multiple time scales is used to estimate both battery SOC and SOH simultaneously,estimating battery SOC at the microscopic scale in real time,and estimating battery SOH at the macroscopic scale after certain conditions are met.Experiments show that the DEKF algorithm has a high accuracy for state estimation.