| With the rapid development of electric vehicles,battery technology,which is one of the three core technologies of electric vehicles,has become a hot spot in the global automotive industry and universities.The estimation of the state of charge(SOC)of the power battery is one of the key technologies of the Battery Management System(BMS).The accurate SOC can not only ensure the safe and reliable operation of the power battery system,optimize the power battery system,Extending the service life is also the basis for the estimation of State Of Health(SOH)and State Of Power(SOP).However,the power battery SOC is a hidden variable and cannot be directly measured,and the lithium ion power battery for the vehicle has strong nonlinearity and coupling,which brings great difficulty to the online estimation of the power battery SOC.To this end,this paper takes the SOC and SOP of lithium-ion battery as the research object,and mainly completed the following research work:(1)The development status of lithium-ion battery for vehicle is introduced,and the offline parameter identification algorithm based on table look-up interpolation and the model parameter identification algorithm based on recursive least squares with forgetting factor are proposed.The research difficulty and deficiency of multi-state joint estimation algorithm for ion power battery is proposed.The SOC-SOP joint estimation algorithm for power battery is proposed.(2)In order to obtain accurate battery status information,a power battery test platform was built to carry out the battery external characteristic test experiment to obtain the input of the SOC estimation model.By analyzing and comparing the error between the output voltage of the n-RC model(n=0,1,2,3,4,5)and the measured voltage,the complexity of the power battery model and the computation time cost,select the appropriate battery model,and establish The online parameter identification model driven by data such as current and voltage in DST cycle condition is compared with the offline parameter identification result obtained by traditional table look-up interpolation method.The results show that the accuracy of offline table interpolation method is lower than that of online parameter identification,and the first-order RC model with online identification method has good prediction accuracy and can be used as the basic model for power battery SOC estimation.(3)For the open loop design algorithm such as Ah integral method and open circuit voltage method,the problem of battery SOC cannot be accurately estimated in real time.A model-based closed loop estimation method for power battery SOC is proposed.The method can not only improve the feedback adjustment precision of the closed-loop system to the initial value of the inaccurate SOC,but also modify the system noise and the observed noise in real time based on the innovation covariance matrix of the motion estimation window,thereby affecting the state variables(SOC,polarization voltage)of the system.and the output variable(terminal voltage).The results show that the model-based dynamic battery SOC closed-loop estimation algorithm has an estimation error of less than 3% and good robustness.(4)Since the conventional power battery peak power estimation result is insufficient in accuracy,and only the transient peak power can be estimated.The transient peak power estimation method based on voltage constraint,SOC constraint and single peak current constraint is proposed.Then the peak power mathematical model under multi-time scale is derived based on the transient peak power,and finally based on adaptive expansion.SOC-SOP joint estimation model based on AEKF(5)In order to verify the estimation accuracy of the dynamic battery SOC estimation model in the real-time system,a BMS semi-physical simulation platform is built and a HIL test model is established.The results show that the model has high output voltage accuracy,which reflects the high SOC estimation accuracy and has important engineering application value. |
PDF Full Text Request