Research On Estimation Of State-of-Charge For Lithium Iron Phosphate Battery

With the growing challenge of energy crisis and environmental problems, lithium iron phosphate battery has been rapidly developed nearly 10 years as a new type of efficient green energy, which is widely used in electronics, electric vehicles and energy storage power supply, etc. In order to improve the safety and service life of the lithium ion phosphate battery, the essentials are effectively control methods and management of the battery. State of charge(SOC) is one of the most important parameters in the battery management system(BMS), which is the foundation functions in the system. Therefore, real-time and accurate estimation of the battery SOC has an fundamental significance.Battery model is the precondition to realize accurate estimation of battery SOC. However, due to its highly nonlinear and complicated working condition, it is difficult to build battery models describing various working characteristics. In order to describe the dynamic and static characteristics of the battery more accurately, this paper have taken hysteresis characteristics between charging and discharging process into consideration. Then, aimed at the disadvantages of the extended kalman filter algorithm, a square root cubature kalman filter algorithm is proposed for SOC estimation, which uses spherical radial rule instead of partial derivative calculation for nonlinear functions. Comparing battery SOC estimation results of different operating condition,it can be proved that the square root cubature kalman filter is better than the extended kalman filter.Finally, with the lithium battery of Tianjin Lishen Company, this paper constructs an experimental platform based on DSP and AT90CAN128. Schematics of related modules have been designed and debugging. The effectiveness and accuracy of algorithm have been verified by runing the SOC estimation algorithm.