Capacity-loss Prediction And State-of-health Estimation In Lithium-ion Batteries

For the echelon use technology of power batteries,as a significant and cuttingedge study of application science,its connotation includes applying the retired batteries that will not meet power performance requirements of electric vehicles any longer to such fields that require lower energy and power density as communication base stations and distributed energy storage through the technical means of the battery mechanism diagnosis,life prediction,state estimation.That means the echelon use of batteries cannot only prolong battery using life and reduce the costs of electric vehicles at the same time,but also avoid the serious environmental pollution caused by waste discharge effectively.However,in existing works,the study of characteristics and aging processes on aging batteries only focuses on the range below 20% of capacity loss,and the related research results cannot be easily deduced to the whole lifetime of the batteries,so that it fails to meet the demand of accurately diagnosing capacity-loss mechanisms and rationally planning the replacement time of battery echelon use.Besides,as the online judgment basis of the replacement time,the existing vehicle-used state of health(SOH)estimation methods are confined to work at room temperature conditions,causing that they do not match the actual working environment of electric vehicles.To solve the problems above,this paper starts with the research on capacity-loss mechanisms diagnosis of lithium-ion batteries,and successively studies the establishment of a whole-life capacity-loss prediction model and the development of an on-line SOH estimation method in electric vehicles,respectively.The main research contents are as follows:First,an improved open-circuit voltage(OCV)aging model,based on nonuniform compression characteristic of electrode potential curves,is proposed to solve the problem that the existing OCV aging models for capacity loss diagnosis gradually distort in the OCV curve fitting process during the middle and late period of the battery aging,leading to the unreliability of the diagnosis conclusions.This model uses the non-uniform compression transformation function between the inherent electrode coordinate system and the available electrode coordinate system,and characterizes the influence of the change of single particle size and multi particle size distribution on the proportion between the single phase region and the two-phase coexistence region of the electrode potential curve,making the OCV aging model keep good fitting accuracy in the whole lifetime.The whole lifetime OCV characteristics experiments,performed on the battery with the current rate of 1C and under temperature 25℃,show that compared with the existing models,the OCV fitting RMS error is reduced from 11 mv to 2 m V by this model throughout the whole battery lifetime.Next,a loss of lithium inventory – loss of active material(LLI-LAM)composite capacity-loss model,based on diffusion induced stress(DIS)distribution theory,is established to solve the problem that the existing capacity-loss prediction models only focus on the change rules of capacity during the early aging stage of the battery so that they cannot provide a reasonable plan for replacement time of battery echelon use throughout the whole battery lifetime.This model,based on the existing LLI model,uses the DIS distribution theory of the spherical particle,establishes a model of material fatigue fracture effects caused by the reciprocating stress during the extraction-insertion lithium process of electrode particles,and characterizes the quantitative relationship between the coupling aging conditions(different temperatures and current rates)and battery LAM rate,making it obtain the ability to predict capacity-loss track of batteries in the whole battery lifetime.The verification experiments,performed on the cells with the current rate of 1C and under the temperature of 40℃ together with the current rate of 0.5C and under the temperature of 25℃,show that compared with the existing models,this presented model in the 5% error tolerance of capacity-loss rate extends the application range of capacity-loss prediction model from less than 20% of capacity loss rate to the whole battery lifetime.Last,an online state-of-health(SOH)estimation method applicable at a wide temperature range for lithium-ion batteries is proposed to aim at the problem that the existing SOH estimation methods are confined to work at room temperature conditions,causing that they do not match the actual working environment of electric vehicles.The method of this paper uses the function relationship between the loss amount of battery capacity and the increase amount of ohmic resistance caused by recycling lithium consumption in solid electrolyte interface(SEI)film formation process,combines the rule model about the effects of temperature change on the resistance values of ohmic resistance components,and breaks temperature range limitation decided by the existing online SOH estimation methods from the aspect of principle.The experimental results show that when the method has the same estimation accuracy as the existing methods(the errors < 5%),it can broaden the temperature range of SOH estimation from 20℃~30℃ to-10℃~50℃.