Modeling And SOC Estimation Of Parallel Battery System Considering Battery Inconsistency

Lithium-ion batteries are widely used in the field of electric vehicle power batteries.Due to their excellent performance,battery modeling and state estimation play a crucial role in battery management systems.In the development process of battery management systems,high-precision models and State of Charge(SOC)estimation are important support for battery system simulation modeling,demand analysis,and charge/discharge control strategies.In this paper,the modeling and state of charge estimation methods for single and parallel battery systems are studied.The main work of this paper is as follows:(1)Taking nickel cobalt aluminate lithium batteries as the main research object,this paper introduces their working principle,battery performance parameters,and analyzes the reasons and mechanisms of battery inconsistency in the battery system.(2)Through parameter identification based on second-order exponential fitting,a battery cell equivalent circuit model based on second-order RC circuit was established.Combined with the characteristics of parallel circuit,a parallel battery system equivalent circuit model was constructed to obtain the SOC basic values of the parallel battery system.The accuracy of the battery cell and battery system models was verified through comparative analysis of simulation results and experimental data.(3)Aiming at the problem of inconsistent current in each branch of a parallel battery system during actual operation,a SOC calibrator considering the inconsistency of the parallel battery system was constructed,and the SOC calibrator was used to obtain a compensation value for the loadΔSOC_C is superposed with the predicted base value SOC_E obtained by the extended Kalman filtering method to obtain the calibrated SOC_r.Using a combination of BP neural network and EKF algorithm to correct the error of EKF algorithm in estimating SOC,a BP-EKF based SOC estimation algorithm for parallel battery systems is proposed.Through comparative analysis of simulation experimental results,the proposed BP-EKF algorithm has good SOC estimation results under constant current discharge,pulse discharge,and dynamic conditions,demonstrating the effectiveness of the method.