Research On SOH Estimation Method Of Electric Vehicle Power Batter

In order to solve the problem of global energy shortage and environmental pollution,and respond to the national policy of peak carbon emissions and carbon neutrality,electric vehicles powered by batteries are gradually replacing fuel vehicles.The Battery Management System is a bridge connecting the power batteries and the vehicle control system.Its performance determines the quality of electric vehicles.As an important function of BMS,State of Health estimation can not only evaluate the remaining useful life,but also provide a basis for the State of Charge and State of Power estimation.At present,the research of SOH still has the problems of difficult modeling and low accuracy.Therefore,a SOH estimation method based on fusion algorithm,ensemble model and multi feature index is proposed.Firstly,the characteristic indexes are selected and the SOH estimation model is established.The working principle of lithium-ion battery,the variation of parameters during charge and discharge,and the internal and external factors affecting SOH decay were analyzed.The actual capacity is selected as the direct performance index of SOH,and Constant Current Charging Time extracted from the charging data is selected as the indirect observation index.By comparing several existing empirical models,a six-parameter ensemble model with higher fitting accuracy is proposed.Further,CCCT and SOH are mapped,and the SOH estimation model based on six parameter ensemble model and mapping function is established.Secondly,the SOH estimation simulation analysis is carried out on the MATLAB.The model parameters are updated and optimized based on Particle Filter,and then the SOH of the battery is estimated.In order to solve the problems of weight degradation and sample shortage of PF,the Improved Coral Reefs Optimization-Particle Filter was proposed.The improved CRO algorithm is used to replace the resampling process and improve the number of effective particles.Based on NASA and CALCE lithium-ion battery data,with the latest observation CCCT as the input and SOH estimation as the output,the effectiveness of this method is verified.Compared with the three existing algorithms,the maximum relative error is reduced by 2.55%and the root mean square error is reduced by 0.91,which verified the accuracy and stability.Finally,the experimental verification of SOH estimation is completed on the independently built experimental platform.Based on the experimental platform,the measurement of SOH and the calculation of CCCT of battery are realized.After parameter fitting and mapping analysis,the estimation model is written into the BMS.Based on the CCCT and ICRO-PF algorithm,the model’s parameters are updated,and then the SOH of No.1 battery after 240 cycles and No.2 battery 740 cycles are estimated.The SOH estimated results are compared with the real value measured directly by the battery tester.The results show that this method can accurately estimate SOH,and the maximum error is less than 6%,which can meet the design needs.