State Of Charge Estimation Of Power Battery Based On Fading Parallel Kalman Filter

The new energy vehicles are a key development direction and strategic emerging industry determined by our country’s “Made in China 2025”.Accelerating the cultivation and development of new energy vehicles is an urgent task for my country to respond to energy and environmental challenges and promote the transformation and upgrading of the traditional automobile industry.Lithium batteries have become the first choice for power batteries for new energy vehicles due to their superior characteristics.State of Charge(SOC),as an indicator of their remaining power,is an important basis for energy management of new energy vehicles.This subject has carried out in-depth research on the accurate estimation of the SOC of lithium-ion batteries.The main work is as follows.Firstly,based on the dual polarization(DP)model,analyze the time-varying characteristics of the parameters in the Recursive Least Squares(RLS)online identification with forgetting factor,isolate the ohmic resistance,and make the RLS identification number of parameters is reduced,the amount of calculation is reduced,and the accuracy of identification is improved.Based on the practicability of offline identification and online identification for different working conditions,an adaptive equivalent circuit model of all working conditions is established,which further improves the accuracy of the model.Simulation experiments show that the full-condition adaptive equivalent circuit model has higher accuracy than the dual-polarization R-DP online model and the DP offline model with separate internal resistance.Secondly,the Adaptive Fading Extended Kalman Filter(AFEKF)is studied,and the error covariance matrix of the traditional extended Kalman filter is weighted by introducing an attenuation factor to reduce the pair of obsolete measurement values.The influence of estimation strengthens the correction function of the new measurement data in filtering,thereby improving the tracking speed and estimation accuracy of the extended Kalman filter.Finally,based on the idea of parallel operation,an adaptive fading parallel extended Kalman filter algorithm(AFPEKF)is proposed to solve the problem of adaptive fading extended Kalman filter in high order.problem.Through the alternate operation of different state variables,the calculation amount of the adaptive fading Kalman filter algorithm is reduced.Theoretical analysis and simulation experiments prove that the AFPEKF algorithm reduces the computational complexity of the algorithm without much impact on the estimation accuracy.