SOH Estimation Method Based On Pure Electric Vehicle Lithium-Ion Battery

With the rapid development of the pure electric vehicle industry,lithium-ion power batteries have also received significant attention as one of its core components.And in order to give full play to power battery performance,improve battery utilization and service life,an excellent battery tube system is essential.At present,some achievements have been made in the research of battery management systems at home and abroad.Among them,the technology of battery state of charge estimation is relatively mature.Compared with this,the research on battery health state estimation is relatively lagging behind,the estimation level still needs to be improved.Therefore,this article takes the ternary lithium-ion battery as the research object,and conducts an in depth study on the battery health estimation by establishing a corresponding lithium-ion battery model.First select the second-order RC equivalent circuit model as the simulated battery model,identify the battery model parameters through the hybrid pulse power characteristic(HPPC)test experiment,and establish the Simulink simulation verification model to verify the battery model under different working conditions,the results of the simulation experiment show that the battery model can accurately describe the operating characteristics of lithium-ion batteries.Then,based on the Dual Extended Kalman filter algorithm(DEKF),an improved algorith based on the idea of finite difference(F-DEKF)is proposed to estimate the battery SOH.The algorithm optimizes the linearization process of the nonlinear system,which is beneficial to reduce the linearization error in the Dual Extended Kalman filter algorithm,and because the algorithm does not involve solving the system Jacobian matrix,it has advantages in terms of model calculation.Finally,the UDDS cycle working condition simulation experiment and the cycle charge and discharge test experiment were carried out for the DEKF algorithm and the F-DEKF algorithm respectively.The results show that the latter has higher estimation accuracy in the Estimation of battery SOH,Compared with the former,the estimation error is reduced by about 0.6%,and the estimation effect is better.