Research On SOC Estimation Of Electric Vehicle Power Battery Management System

In recent years, with the vigorous development of the world economy, the electric vehicle has already become the new emerging research object of automobile industry.Combined with the global energy crisis and environmental issues, the scale and influence of electric vehicles are gradually catching up with traditional fuel vehicles. Nowdays,lithium iron phosphate power battery as the mainstream power supply of the electric automobile, its advantages is beyond doubt. Among them, the safe and reliable battery management system is the key; its main function is to monitor the vehicle power battery and estimate the State Of Charge(SOC). In this way not only can scientifically manage the work status of the battery and prolong its service life, but also can provide accurate information of battery status. According to these information, the drivers can reasonably plan the trip and charging time. Therefore, establish a reliable battery management system and accurate SOC estimation algorithm has the very important significance of improve the performance of electric vehicle.In this paper, iron phosphate lithium battery as the main object of study, according to its structure diagram, detailed described its working principle and characteristics. And on the basis of the battery charge and discharge experiments, established the Thevenin equivalent circuit model and identified of its dynamic parameters.Finally, used the experiment of pulse discharge verified the accuracy of the battery model.And researched on battery SOC estimation algorithm, introduced the working principles and advantages and disadvantages of several traditional estimation algorithms,detailed analysis the working performance of the Extended Kalman Filter(EKF) algorithm and applied to the SOC estimation of the battery. On this basis, to further reduce the estimation error of EKF algorithm, this paper introduced the BP(Back-Propagation)artificial neural network to optimize and remedy the estimation results of the EKF.Buildded a SOC estimation simulation model based on the improved BP-EKF algorithm,and compared and proved the estimation accuracy of the BP-EKF& EKF algorithm under various conditions.Finally, this paper designed the Battery Management System(BMS) platform and offered the topology map; completed on BMS hardware platform based on use DSP(TMS320LF2407) as the main control chip. For the composition and main function of the BMS platform hardware, introduced work principle and process of the SOC estimation software system, and eventually realized the accurate estimation of SOC in the battery experimental platform. The experimental results shown that, using the improved algorithm could improve the estimation accuracy greatly, this not only reflects the reliability of the experimental platform, but also verifies the correctness and validity of the improved algorithm.