Research On State Of Health Estimation And Retirement Sorting Methods Of Li-ion Batteries

As a key monitoring indicator in the battery management system,the state of health(SOH)of the lithium-ion battery reflects the aging performance.When the battery SOH is lower than 80%,it needs to be retired according to regulations,and the retired batteries can be sorted and applied to other suitable occasions to improve resource utilization.In order to solve the problems of low battery SOH estimation accuracy and long retirement sorting time,the main research contents of this thesis are as follows:Firstly,in order to solve the problems of difficult selection of hyperparameters and low estimation accuracy of lithium-ion battery state of health estimation model,a state of health estimation method based on improved long short-term memory neural network is proposed.According to the data during the battery charging and discharging process,the aging features reflecting the battery’s SOH are extracted,and the grey correlation method is used to quantitatively analyze the correlation between the features and the SOH.On this basis,with the selected five aging features as model input and the SOH as the model output,the long-short-term memory neural network algorithm is used to establish the SOH estimation model,and the particle swarm algorithm is used to optimize the key hyperparameters in the model.The experimental results show that the proposed method achieves accurate estimation of the state of health of lithium-ion batteries.Secondly,in order to solve the problems of long sorting time and low classification accuracy of retired lithium-ion batteries,a sorting method based on multi-feature parameter fusion is proposed.Firstly,to realize the fast and accurate estimation of the remaining capacity of the battery,the incremental capacity curve is obtained according to the constant current charging process of the battery,and the curve features are extracted,and then the relevance vector machine algorithm is used to establish the capacity estimation model.After obtaining the remaining capacity of the battery,the fuzzy C-means clustering algorithm is used to preliminarily sort the retired batteries.For batteries with similar remaining capacity,to improve the consistency of battery capacity decay during subsequent use,a further sorting method based on internal resistance and polarization characteristic parameters is proposed.Choose to conduct HPPC test on the battery under the state of charge of 10%.Based on the second-order RC circuit model,use the recursive least squares algorithm with forgetting factor to identify the model parameters,and then calculate the comprehensive evaluation value of the battery.Further sorting is done according to the distance of the comprehensive evaluation value of the battery.Finally,the effectiveness of the proposed battery state-of-health estimation method and retirement sorting method is verified using the lithium battery cycle aging experimental dataset and the retired lithium-ion battery experimental platform.