Lithium-Ion Battery SOC Estimation System with CAN Bus Communication

• Main Authors: Yu-Shuan Lien, 連于瑄
• Other Authors: Yi-Hua Liu
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/rsezz6

碩士 === 國立臺灣科技大學 === 電機工程系 === 104 === In this thesis, a modularized SOC estimation system for series-connected lithium-ion battery pack is proposed. The presented SOC Estimation system provides some features, including cell voltages, pack temperature and pack current monitoring, calculation of battery state of charge (SOC) and remaining capacity and Battery overcharge/over discharge protection. In this thesis, an integrated SOC estimation method which combines coulomb counting method and open circuit voltage (OCV) method is presented. During charging/discharging modes, coulomb counting method is exploited to obtain the charged/discharged capacity by integrating the in-and-out-flowing current over time. On the other hand, OCV method is utilized in the relaxation mode to estimate the SOC according to the OCV and SOC relationship. If the current during relaxation mode is non-zero, an additional I*R drop correction will also be employed to achieve better accuracy. In the proposed system, a master/slave configuration is utilized. The proposed BMS has an on-board CAN bus interface which allows it to communicate with other CAN BMS modules. In this thesis, the central controller utilized in the master/slave BMS is digital signal controllers (DSCs) dsPIC33FJ64GP802 and dsPIC33F16GS502 from Microchip Corp, respectively. In order to provide the real-time display and data logging of all the battery parameters, a user-friendly graphical user interface (GUI) is also developed using LabView from the National Instrument Corp. Detailed description of the hardware and software of the proposed SOC estimation system will be provided, and experiments will also be carried out to verify the correctness of the proposed SOC estimation system. According to the experimental results, the accuracy of the voltage/current measurement is higher than 99.94 % and 99.77 %, respectively. The maximum error of SOC estimation is lower than 3 % and the maximum error of estimated run-time is fewer than 2 minutes.