Research On Power LiFePO₄ Battery Management System

Facing with the increasingly prominent contradiction between supply and demand of fuel and environmental pollution problems, the world's leading automobile producers have accelerated the pace of research and development of new energy vehicles. Electric Vehicles?EV? have become the developmental direction of automobile industry. This trend shows that the battery electric vehicle power device is essential, and LiFePO₄ battery with superior performance has become the first choice of power electric cars. As the power device of electric vehicles, high requirements have been put forward for its safety, cost and endurance etc. As one of the key technologies, Battery Management System?BMS? is committed to improve application security, reliability and efficiency of power battery.This dissertation chose the 50 Ah LiFePO₄ battery composing of 8 single cell as the research object, analyzing the mechanism of performance testing. The dissertation designed a set of battery management system with DSP as the core to monitor the battery working state, improved the estimation accuracy for Battery State of Charge?SOC? and enhanced the security and reliability of the battery.Firstly, through consulting a large number of literatures, the research status of BMS was analyzed and summarized, the characteristics and working principle of LiFePO₄ batteries were in-depth studied. In addition, by comparing the advantages and disadvantages of existing SOC estimation methods, an improved PSO-BP neural network for SOC estimation was proposed to improve the accuracy of SOC by optimizing the original BP neural network.Secondly, using TMS320F2812 as the control chip, the hardware and software circuits of battery management system were designed. According to the system requirements, the hardware circuits include the battery working state information detection circuit, communication circuit, a balanced control circuit and protection circuit, etc. Meanwhile, all system software designs were completed, including the main program design, the battery information detection subroutine design, CAN communication subroutine design, SOC estimation subroutine design, etc.Finally, the overall debugging of battery management system was completed in this dissertation. By setting up the test platform for BMS, signal acquisition and communication functions of BMS were verified. Though designing the experiment, the test for balanced control and SOC estimation function of BMS were finished. Experimental results show that the signal acquisition accuracy of BMS satisfy the design requirements in this dissertation. BMS reveals effectively communication between the parts, good performances of balance control and SOC estimation.