Research And Application Of Active Equalization Technology In Battery Management System

With the dwindling oil reserves and the increasingly serious vehicles exhaust pollution,countries around the world are in urgent need of finding new directions for future vehicle development.In this context,new energy vehicles have gradually become the focus of vehicle development in various countries because of their ability to effectively reduce energy consumption and reduce air pollution.Lithium-ion battery has become one of the three core components of new energy vehicles because of its high energy density and no memory effect.In practical applications,in order to meet the needs of different available capacities and different voltage levels,it is usually necessary to connect individual batteries in series into groups.However,subtle differences in manufacturing processes and different operating environments will continue to exacerbate the inconsistencies that already exist between individual cells,which will lead to a decreasing cycle life and increasing risk of battery pack use.In order to slow down the decay of battery pack cycle life and avoid safety hazards during electric vehicle(EV)driving,it is especially important to incorporate battery management system in EVs,among which State of Charge(SOC)estimation and equalization control are two critical aspects.To address the problems of poor stability and weak tracking effect of the Extended Kalman filter(EKF)algorithm and the problems of long equalization route and single equalization path of the single-layer equalization topology,this thesis conducts research on both improving the EKF algorithm and the equalization topology.The main research contents and conclusions are as follows.(1)In this thesis,the complex electrochemical structure of the battery is simulated by a second-order RC circuit with high model accuracy and moderate computational effort to establish an equivalent circuit model.On this basis,the SOC estimation process based on the modified error covariance extended Kalman filter with fading factor(FF-MCEKF)algorithm is established by combining the extended Kalman filter algorithm and two covariance correction methods,and the simulation is verified in Matlab based on constant current static operating conditions and DST dynamic operating conditions.The results show that the improved algorithm can avoid the appearance of sick covariance matrix and prevent filter divergence.Meanwhile,the estimation error is reduced by 26% compared with the EKF algorithm.(2)By analyzing the research status of battery equalization technology and equalization control strategy at home and abroad,a hierarchical equalization topology based on a bidirectional flyback converter and a multiswitch inductive LC converter is proposed to improve the disadvantages of the large size and high cost of the bidirectional flyback converter and the single equalization path and slow equalization speed over long distances of the multiswitch inductive LC converter.Based on this,different equalization control strategies are developed for the two equalization structures,and the equalization topologies and equalization control strategies are fused and verified in Matlab/Simulink.The results show that the equalization topology proposed in this thesis can equalize the power of each individual cell in the battery pack to the same level and maximize the overall available capacity of the battery pack.(3)The equalization system software and hardware platform is built on the basis of 8 series-connected batteries,and the equalization control strategy is invoked to equalize the battery pack.The results show that the SOC of each individual battery converges after 14min20 s under charging,and converges after 27 min under discharging.It proves that the equalization circuit designed in this thesis has the ability to improve the inconsistency of the battery pack.