THE STATE OF CHARGE ESTIMATION FOR LI-ION BATTERY ON CHARGE AND DISCHARGE STATE BY EXTENDED KALMAN FILTER

• Main Authors: Teng-yao Chang, 張登堯
• Other Authors: Chung-chun Kung
• Format: Others
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/88q2n9

碩士 === 大同大學 === 電機工程學系(所) === 102 === To increase the life of the lithium-ion battery,and to improvethe safety andefficiency of the battery used, the accurate determination the battery's capacity is very important. Because of the complexity chemical and physical processes, the battery information that we can use is fewer and fewer. So estimation the state of charge is very difficult. In order to determine the state of charge, open-circuit voltage and battery's internal resistance are the important parameters for estimation the state of charge. Accurate estimation the state of charge can avoid over charging and discharging, improve battery performance and extend battery life. However, for batteries, the state during charging and discharging are not the same, this thesis using the equivalent circuit model(ECM) to approximate characteristic of the Li-ion battery. In order to obtain the unknown parameters of the ECM, the open circuit voltage (OCV) test and direct current internal resistance (DCIR)testare used to find the relationship between the parameters and SOC of the Li-ion battery.From the experiments results, we can find the relationship between the parameters and SOC of the Li-ion battery is a nonlinear, and the initial values will affect the accuracy of the SOC estimation. These problems can be effectively improved by the extended kalman filtering. Therefore, in this thesis we use the extended kalman filtering(EKF) to estimate the SOC of the Li-ion battery, andreduce the impact of SOC estimation by parameters errors and the dependence of the initial values. Finallythrough simulations and experimental results to validate the accuracy for ECM, and confirm the effective of EKF.