Development of an Innovated Battery Management System for a Heavy Electric Scooter

• Main Authors: Chung-Yung Wu, 伍崇永
• Other Authors: Lin, Jeng-Chyan Muti
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/69480531896528058740

碩士 === 國立勤益科技大學 === 電機工程系 === 102 === At present, electric vehicles are considered one strong candidate to reduce oil reliance for transportation sector. Electrical vehicles have also become one important too for countries to reach goals of saving energy and reducing carbon dioxide emissions. Electric vehicles have the benefits of zero emission, low noise and easy operation. Also, Electric vehicles can reduce air pollution and noise to our dwelling environment. Lithium batteries are the most widely used in electric vehicles, because they have higher energy and power densities, lower self-discharge rate and zero environmental pollution. Among various lithium batteries, lithium iron cells (LFP) are well adopted by EV industry since LFPs are immune from self-ignition and explosion when batteries are over-charged or over-discharged. LFPs are safer storage battery when compare with other lithium batteries so they are more suitable for electric vehicles. Due to different internal resistances or aging processes, cells of LFP exhibit inconsistencies in capacities and discharging characteristics. These inconsistencies cause early overcharging or over discharging when inconsistent cells are connected in series. Early overcharging or discharging results in the reduction of the battery capacity. If not properly protected, inconsistency might lead to performance deterioration. It is therefore imperative to have a battery management system to keep the battery within safe operation boundary. The major tasks of the battery management system are cell voltage measurement; current measurement; temperature measurement, cell protection of overcharging, over discharging, over temperature and over current; state of capacity (SOC) prediction. In additionally, cell balance by various schemes is also necessary in the BMS to overcome cell inconsistency problem. This thesis designs an advanced battery management to be applied to a high performance electric motorcycle. LFP battery of 72V or 24 cells in series is adopted for the current study. The developed BMS is aimed to improve the shortcomings in existed BMSs. Special emphases are as followings: a new cell balance scheme based on a globally active balance circuit to compensate cell inconsistency; a split type BMS which separate balance module with main control board to save space and to reduce heat loadings in the battery pack; a hybrid state of charge estimation scheme for provide reliable SOC estimation to drivers. BMS with built in balance module usually need extra space and heat dissipation mechanisms to accommodate the balance module. It therefore better to have a split type BMS where balance module integrates with charger and not included in the BMS with battery pack. This would save space and cost for the battery module. This study separates BMS into two subsystems as 72V mainly control board and 72V balance board. 72V main control board is installed at electric vehicle and 72V balance board is integrated with the charger. 72V main control board can measure cell voltages of 24 cells in series with equal high accuracy in every cell. The BMS is based on PIC18F87K22 microcontroller of Microchip and main tasks of the BMS are to measure cell voltages, currents, temperatures, vehicle speed and also to protect battery from over-charging, over-discharging, over-current, over-temperature. SOC prediction, LCM display and data storage for post analysis are also executed by the BMS. The BMS also conducts two-stage charging control, which is a unique feature of the current study. We build an energy management system to control and data acquisition of the performance data of the electric motorcycle and battery. At discharging state, the LCM monitor displays the electric parameters. At charging state, the system transmits message to Labview HMI by RS-485 communication. The parameters are analyzed and calculated by the energy management system and displays on HMI. In order to test performance of the developed BMS, we utilize self-made electric scooter as test platform. This scooter is made of steel and used a 72V rear hub motor. The electric scooter is tested on city pavements. According to experiments, operating power of the motorcycle is 3.9kw, average speed is 40km/h, maximum speed is 75 km/h and mileage is 111.2 km. The result shows that three-stage SOC method is 95% accuracy.