Development Of Battery Management System For Pure Electric Vehicles

The battery management system for electric vehicles are in charge of battery state monitoring,battery overcharging and overdischarging preventing,and battery life extending.By improving the accuracy of estimated SOC state of power battery,the power battery can get a better performance in electric vehicles.In this thesis,the scale of influence to the discharge capacity of the battery,which caused by the voltage,current and temperature,was obtained through experimental testing of ternary lithium battery.Meanwhile,this thesis analyzes the advantages and disadvantages of different methods of estimating SOC state of power battery.Based on the adjusted and improved RVM,I proposes three SOC estimation methods which cover RC-RVM,AG-RVM and AG-RC-RVM.I find out the special SOC estimation method of ternary lithium battery,which based on AG-RC-RVM,has the characteristics of high accuracy,strong capability of fitting and generalization,and high speed of operating.Based on demanding,I analyze the function of battery management system with technical solutions,and construct the overall master-slave frame of battery management system.Depending on the main controller MC9S12DT128,I design the main control system hardware circuit,which mainly includes power supply,clock,total voltage acquisition,total current acquisition,SOC state estimation,insulation detection,CAN communication and thermal management modules.At the same time,based on the main controller MC9S12P64,I design the slave control system hardware circuit,which mainly includes power supply,temperature detection,voltage acquisition and equalization,and CAN communication modules.I use language-C as developing language.Based on the concept of modular software design,I design the main program,system initialization program and functional module subprogram,which are focusing on the main control system and the slave control system separately.This thesis executes functional test verification on the developed battery management system,including cell voltage acquisition and equalization,temperature acquisition,insulation detection,total voltage acquisition,total current acquisition,SOC state estimation,thermal management,and communications.In this thesis,I develop a battery management system for pure electric vehicles,which estimate the state of SOC by data learning method.Implementing this developed system,the accuracy of estimated state of SOC is high.At the same time,this paper also has significance of reference for the practical research and development of the battery management system.