The Design Of Distributed Battery Management System For Electric Vehicles

With the progressing in electric vehicle technology and the increasing of low-carbon green life philosophy, electric cars are going into people’s daily lives depending on lots of advantages, which will brings great convenience and reduce the environmental pollution while we are driving them for moving. That gets widely praised. Our country is trying to promote the development of new energy vehicles and that caused positive reaction, at the same time many companies are also competing to join the field to research and develop electric vehicles. It can be said that the electric car has entered a rapid development period, which will not only bring new economic growth point and be beneficial to the environment, but also contributed to the battery management system development. Electric vehicle battery management system as an integral part of the car can effectively improve the mileage of electric cars and battery life while reducing the cost.To improve usability and to facilitate expansion, the battery management system is a distributed architecture of master-salve. The system consists of a master module for processing task and some salve module for gathering information of the battery. Both of them can send and receive information and commands via CAN bus. The number of salve modules number can be 1 to 127 and all of them can be assigned a separate CAN bus source address. With the LTC6804 battery monitoring chip, salve modules have high-precision voltage acquisition. The master module hardware circuit using high-precision voltage divider resistors and 16-bit AD functional module to improve the precision of voltage and all the hardware consider the anti-jamming performance to ensure that the system is stable and reliable.The battery management system to achieve efficient management for the batteries needs to base on accurate battery models. This paper used some lithium iron phosphate battery to do charge and discharge experiment and get lots of experimental data. This paper used variable parameters of capacitance and resistance as well as Thevenin equivalent circuit to identify model parameter with the help of battery data and to demonstrate the improved results by using open circuit voltage method.State of charge estimation as one of the core function of the battery management system has important significance. This paper based on improved battery model as well as extended Kalman filter to study SOC estimation and use finite difference method to instead of classic algorithm to improve SOC estimation. This paper gave some comparison to verify the improvement effect and that will apply to the battery management system.The last chapter is about of the battery management software system design and testing. The software system used hierarchical and modular design method, which divided into bottom, middle layers and application layers. Besides, the master module software intermediate layer was added uCOS-II System to promote real-time response. In this paper, the main program module, battery management module, CAN communication module and the charging module programming have been described. By reading and interpretation of CAN messages can get the system information. At last, Some test result are given for the achieved function and this paper analyzed the results and draw the conclusions.