The Research On SoC Estimation Method And Evaluation Of Lithium-ion Battery Pack For Electric Vehicle

The State of Charge(SoC)is the core of electric vehicle battery management system.Its accuracy directly affects battery life,functional safety,balance control and thermal management strategy.Signal noise interference,battery model adaptation to temperature and aging,and cell inconsistency are important factors that affect the accuracy of battery pack SoC estimation.In order to achieve accurate estimation of lithium-ion battery pack SoC,a method is suggested in this paper by combining the Interacting Multiple Model(IMM)and Adaptive Battery State Estimator(ABSE),with 18650 lithium-ion battery as the research object.the main work of the paper is as follows:(1)the working principle and model classification of lithium-ion battery are studied.According to the characteristics of working condition,the equivalent circuit model composed of slow-varying voltage and total internal resistance is selected in constant current and constant-voltage conditions,and the first-order RC model was selected for the dynamic discharge conditions.The effects of charge-discharge rate,temperature and inconsistency on the open-circuit voltage,capacity and internal resistance of the lithium-ion battery were analyzed.(2)based on the comparison and analysis of traditional parameter testing process and parameter identification accuracy,the battery parameter testing process and parameter identification combination are optimized to obtain battery model parameters.A mathematical model of the slow-varying voltage,total internal resistance and open circuit voltage is established to obtain an equivalent circuit model that can be applied to multiple working conditions and the whole climate.(3)based on the comprehensive charge and discharge characteristics,the battery module equivalent circuit model is established.The IMM-ABSE algorithm is proposed,and the SoC is estimated by ABSE and embedded in the IMM model.The SoC of each model is probabilistically fused according to the information distribution factor to obtain a battery pack with higher precision.The accuracy and stability of the AH integration method,the average voltage method and delta SoC method based on the unscented Kalman filter and the IMM-ABSE algorithm are compared.The robustness and universality of the proposed algorithm are evaluated under different temperature conditions.The experimental results show that the method is suitable for the environment where the input signal is noisy,the whole climate is multi-working and the battery is inconsistent.The error during the effective charge and discharge is stable within 2%.(4)generating the target C code of IMM-ABSE algorithm through Simulink automatic code generation technology,and downloads it to the BMS controller for bench and real vehicle verification.And the development of lithium-ion battery pack monitoring and state estimation PC software for real-time recording and algorithm verification.The experimental results show that the IMM-ABSE algorithm based on the actual BMS controller results consistent with MATLAB simulation results.At the same time,it has good robustness and stability under the condition of vehicle,and there is no obvious fluctuation during driving,which basically meets the requirements of practical applications.