Research On Key Technology And Test System Construction Of Battery Management System

Battery management system is one of the most important management systems in electric vehicles.It is related to the service life,cost and safety of power batteries.Therefore,the study of battery management system has theoretical significance and application value.Power battery is a time-varying nonlinear control object,which will be affected by the working condition,environment and other random factors in the application of electric vehicles,which makes the real-time accurate management of power battery more challenging for the battery management system.In this paper,the specific research work for the battery management system of electric vehicles includes:(1)Firstly,the power battery and battery pack test platform were built,and the battery parameter identification data was obtained by improving the battery power pulse test method.Then,based on ADVISOR,a complete vehicle model of the battery application environment was established to obtain multiple condition test data of the power battery for the application vehicle,and the battery model was verified by using the battery dynamic condition data.Finally,the relationship between available capacity of power battery and discharge ratio,temperature,aging condition and self-discharge is studied to guide the identification and update of battery parameters in the later stage and complete the accurate modeling of power battery.(2)Adaptive matching and comparison were made between different power battery models and model parameter identification data,and the battery identification data adapted by different power battery equivalent circuit models were compared,so as to realize the tradeoff between the complexity of power battery model and identification accuracy.By analyzing the factors affecting the accuracy of the battery model,the battery model was improved,and the influence of the current,temperature and state of charge for the power battery parameters was considered.Finally,a general model off-line parameter identification method based on Simulink parameter estimation tool is proposed.(3)Two high voltage acquisition methods in current battery management system are fused to solve the error and noise problem of total voltage acquisition data.The fusion model and Kalman filter are improved to reduce the computational complexity of data fusion.Finally,it is verified that the proposed method has high accuracy and stability.In order to determine the SOC and SOP of the power battery,an online parameter identification method based on data driving was proposed,which captured the characteristics of the battery in real time and updated the model parameters with the recursive least square method with forgetting factor.Then,a multi-state joint estimation algorithm based on adaptive extended Kalman filter is designed,and a mathematical model of continuous peak power with multiple constraints and sampling interval including voltage,residual charge and peak current of single unit is established.Finally,the simulation model based on the actual operation data of pure electric vehicles was built in Matlab/Simulink environment.(4)In order to make the battery pack to play the best performance and life,it is necessary to carry out thermal management of the battery and control the battery temperature in a reasonable range.In this paper,combined with the research group on phase change materials,the practical application of phase change materials in battery packs was improved,and the simulation and verification of the battery thermal management scheme based on the combination of 3D honeycomb composite phase change materials and liquid cooling were carried out.(5)To achieve requirements validation based on executable models,iterative simulation design,automatic code generation,and continuous testing and validation.In this paper,the advanced model design process is used to build the BMS controller and the tested object model,implement the model in the loop to test various functions about BMS,and the charging communication process of BMS is verified by d SPACE.Finally,the hardware-in-the-loop test platform based on RT?LAB is built to complete the comprehensive test and verification of the BMS.In this paper,the high voltage acquisition algorithm based on multi-sensor fusion technology,dynamic continuous peak power estimation algorithm with multiparameter constraint and battery thermal management scheme based on phase change material and liquid cooling coupling are proposed,which solve the main problems existing in current battery management system,such as parameter acquisition,state estimation and battery temperature control.This paper proposes a general battery model parameter identification method and an improved battery model considering current,which solves the tradeoff between complexity and precision of power battery model,and provides a theoretical basis for the development of battery management system for electric vehicles,and has good practical application value.