Research And Design On Lithium Battery Management System For Pure Electric Vehicle

Energy shortages and environmental pollution are two outstanding problems the humanity faced. In recent years, the traditional automotive industry has been developing rapidly,total vehicle presents a stepwise increasing trend. In the service of the economic growth and people’s living needs, the traditional automobile industry also caused serious environmental pollution and resource consumption. The pure electric vehicles, by advantages of non-polluting and highperformance, are becoming into the development direction of the contemporary car. Power lithium battery with a very prominent energy storage and dynamic performance, is considered the most suitable pure electric vehicle power unit. However, if there is no excellent battery management technology, battery performance after a group will be much lower than a single cell. Electric cars will be exhibited with short journey, low security, the high cost of battery replacement and other issues. Therefore, pure electric vehicles in order to get market recognition, we must overcome these problems,improve the competitiveness of this new electric vehicle products.This paper analyzes the basic operating characteristics of lithium batteries,and focus on the battery state of charge(SOC) estimation, balance management, maximum power prediction,insulation management. Lithium battery SOC estimation is currently the bottleneck and emphasis of battery management system(BMS) research. This paper first to Thevenin model for the system model, which describes how to use the extended Kalman filter to estimate the SOC of lithium batteries, And in accordance with its inherent defect and operating characteristics of lithium battery,it proposed a new method for estimating SOC based Dual model adaptive extended Kalman filter. Balance management can effectively improve the inconsistency of lithium batteries, the current balance management system often as the terminal voltage of the lithium battery equalization object, through the balance circuit to make the terminal voltage of the battery to in consistent. However, the consistency of the voltage does not really achieve to increase batteries capacity utilization. Therefore, on the basis of the flyback transformer topology as the equalization circuit, this paper by maintaining the consistency of the lithium battery SOC, to achieve a real sense of balance between the batteries pack.In this paper, the lithium battery management system for pure electric vehicles uses a combination of centralized and decentralized hardware structure. The local electric control unit is TMS320F2812 DSP, it is responsible for the main work of the entire battery management system, including SOC estimation, balance management, safety management, data processing, troubleshooting, the total voltage and current measuring. The underlying control level consists of Atmega32 MCU, each MCU is responsible for collecting eight battery voltage and temperature, as well as controlling the balance unit. The MCU and DSP uses CAN bus for communication, between the DSP and vehicle control systems and charger also communicate via CAN bus. In this paper, the system software design using modular, structured way, the main program as well as the functional modules of the system is designed.Finally, the lithium battery model and SOC estimation method is verified by simulation,and the system of data acquisition function and balance function was tested. Simulation and test results show that the test circuits and estimation algorithms meet the design requirements.