Study On Sorting Technology For Lithium-ion Battery Of Electric Vehicle

Cell separation technology is to use some technical means to screen out the different batteries in terms of performance,so that the rest similar cells can be easily recombined.By cell sorting and regrouping,the service life of the battery pack can be greatly extended.The RC equivalent circuit battery model is built according to the basic principle of circuit and the resistance parameter can be obtained using the charging date of the monomer battery by least-squares fitting methods.According to the resistance data,the SOH of battery can be determined.Monomer battery charging voltage curve is clustered according to the curve similarity based on dynamic time warping distance,so the battery types can be determined.Meanwhile,Elman neural network model is built for classification to improve the accuracy of classification and futher reduce the classification error caused by interference and other factors.The charging voltage curve of the baseline and charging voltage curve similarity can be obtained.The similarity of charge voltage curve and the SOH are the input and the category of the battery is the output.Elman neural network will learn and self-optimization with the input-output learning samples.In order to improve the learning speed and easy to fall into the local minimum,Genetic Algorithm Elman neural network model is established.Finally,based on the SOH and the charge voltage curves cluster result,the Elman neural network is optimized.The result provides guidelines for the combination and replacement of Electrical vehicles Lithium-ion batteries.The simulation results show that the curve clustering method based on the charge voltage curve can effectively distinguish the inconsistency of the lithium-ion battery of the electric vehicle,so the improved Elman neural network has superior performance and the sorting method of Lithium-ion battery has a fine effect.