Research And Implementation Of Lithium Lon Battery Management System In AGV

Automated Guided Vehicle (AGV) is an unmanned carrier, which is equipped with battery as power source and automatic guidance system. At present, AGV mainly uses lead-acid batteries and nickel-cadmium batteries to supply power. In the case of lead-acid battery, its energy density and the charge&discharge rate are low, besides, the battery life is short. As for the nickel-cadmium battery, it leads to heavy pollution, and it needs regular maintenance. In comparing, the lithium titanate batteries have se-rious of advantages, like high charge&discharge rate, high energy density, better secu-rity, long service life for more recycling times and no pollution, etc. So it can replace lead-acid batteries and nickel-cadmium batteries to act as the power of the AGV. To ensure the safety of the lithium titanate batteries during working process and extend the battery life, it is of great significance to develop a Battery Management System (BMS) according to the the AGV specific needs.First of all, an in-depth study on the performance characteristics of lithium titanate batteries has been made in this paper. A series of charging tests and discharging tests have been performed at various temperatures and different charge/discharge currents. Through the experiment, the influence of temperature and current rate on certain facet of the battery (the charge and discharge efficiency, open circuit voltage, the capacity) has been researched. Second, the no n-adaptability of the existing State of Charge (SOC) definition has been pointed out and a renewed definition of SOC has been given. It de-picts the SOC from a different perspective by introducing the absolute charge state, the static state of charge, the dynamic state of charge. Then we estimate the new SOC of the battery with Extended Kalman Filter (EKF) algorithm, and has a simulation of the algorithm with Matlab and then improves the EKF algorithm. The simulation results show that the improved algorithm is suitable for a variety of conditions, and the esti-mation error is within±3%. Finally, hardware and software of the BMS, which can sa- tisfy the specific requirements of the AGV, has been designed. The main chip in the hardware system is MC9S12XET256microcontroller. The chip LTC6802is used to to collect data and manage passive equalization of each cells. The software includes a PC software to monitor BMS and a embedded software which schedules applications with μC/OSII operating system. Comprehensive tests of functions had been made on the de-signed BMS. Ultimately, the results showed that the performances of the BMS achieve a predetermined value.