LiFePO₄Battery Pack Energy Management System Research And Design

As the most potential energy storage battery, the LiFePO4battery has a great commercial development value. The LiFePO4battery is low-cost, safe, good on cycling and large on capacity. Because of the variance of material and skill during manufacturing the batteries, a great difference to the SOC estimation of each battery and the battery pack would occur. The variance will reduce the life of battery pack with long time running, so it is important to build a management system to diminish the influence. This thesis focuses on energy managing of the LiFePO4battery pack.Through comparing kinds of different SOC prediction methods, we use the Extended Kalman filtering methods to predict the SOC of the LiFePO4battery, the advantages and practicability of the Kalman filtering method are discussed. In order to implement Extended Kalman filtering method on our project, the calculation and analysis of the method are studied to make sure the accuracy of the theory, the formula are deduced and operation step of Extended Kalman filtering method are listed.In order to simulate the external characteristics of the LiFePO4battery, we built the optimal equivalent model of the LiFePO4battery based on the experimental data of the external characteristics of the LiFePO4battery, such as terminal voltage, current, etc. The electrical vehicle experiments are designed to get the data, and two LP44147132AB LiFePO4battery packs are used to supply the power to the vehicle in the experiment. The vehicle ran on the flat road, and the laptop measured data of each battery in the battery pack, including time, voltage and current. The open circuit voltage(OCV) vs current of Single LiFePO4battery was worked out by these data. The equivalent circuit model of the LiFePO4battery is built by MATLAB. The OCV and current are as reference data, the variances are figured out by fitting simulation data to experimental data, and the fitting methods are Extended Kalman filtering and least square. These variances and the equivalent circuit reflect the real battery statue well. The active balancing method was used to balance the batteries in pack. The battery management module is designed based on this method. The LiFePO4battery equivalent was used to estimate the battery SOC, and the battery management system was built after the software and hardware were cooperated. The experiment implemented with balancing circuit module and without balancing circuit module separately, then the conclusion is draw and analyzed. The experiment implemented with balancing circuit module and without balancing circuit module separately, then the conclusion is draw and analyzed.The contrast experiment shows that with balancing circuit module the system have a long time running than the without one, more power output and closer output and input character. The coefficient of utilization of the controlled battery is improved and the battery output power increase.