Research On State Of Health Estimation Of Lithium-ion Battery And Screening Method Of Retired Battery

As a core component of energy storage system,lithium-ion battery has been widely applied in many industrial fields due to its superior performance.In the process of using lithium-ion battery,the degradation of battery performance may affect the normal operation of the whole system.In order to ensure the safety and stability of the whole system,it is necessary to carry out real-time monitoring of the battery health status.However,in practical applications,there exist difficulties in measuring online the parameters directly representing the aging degree of batteries,such as battery capacity and internal resistance.In addition,battery needs to be decommissioned when its performance cannot meet the application requirements.The secondary utilization of retired batteries can avoid resource waste and environmental pollution.The stringent screening of battery can reduce the inconsistency of reassembled batteries,which is an important part of the secondary utilization.However,the existing screening methods are cumbersome and inefficient,which make it difficult to be applied to the mass screening of retired batteries.For purpose of solving the above problems,the main research work of this paper is as follows:First,health features indirectly reflecting battery performance are studied and a health feature extraction and optimization method based on partial charging curve is proposed.The appropriate charging curve interval is selected according to the analysis of battery aging characteristics.From the charging voltage,current and temperature curve in the selected interval,10 health features are picked out and canonical correlation analysis is utilized to optimized the health features.As the result of optimization,one-dimensional fused health feature(FHF)has the highest correlation with capacity,which construct a low-dimensional health feature that has potential to adapt to complex working conditions.Second,according to research on health diagnosis methods of lithium-ion battery,a joint state of health(SOH)and remaining useful life(RUL)estimation framework based on health feature optimization is presented.As the model input,the FHF obtained is applied to establish a capacity degradation model based on gaussian process regression.Furthermore,least squares support vector machine is used to predict FHF in the future cycles,and then the FHF predicted are combined with the established battery aging model to realize RUL estimation.The experiments of different operation conditions and battery types show that the proposed joint SOH and RUL estimation method has high accuracy and reliability.Finally,on the basis of the analysis of retired battery characteristics,this paper proposes a retired battery screening method based on battery historical usage data.The performance evaluation indexes include SOH,RUL and internal resistance.The analytic hierarchy process is utilized to give evaluation indexes different weights in line with the importance degree of each index.Then,retired batteries are classified based on the grey correlation degree and the threshold value which is set according to application requirements.To improve the performance consistency between batteries,the fuzzy c-means clustering algorithm is used to conduct battery screening for the same kind.The validity of the proposed method is verified by sorting and reorganizing 200 retired batteries.