本文提出一種應用於鋰離子電池的電量狀態(state of charge, SOC)估測方法,利用等效電路模組(equivalent-circuit model, ECM)來表示鋰離子電池電化學反應,而等效電路中的參數則是使用GBT-2211電池量 測平臺對電池進行充放電測試所擷取電壓、電流等數據後以最小二乘法(generalized least squares, GLS)辨識建立,其中欲辨識參數與變數間之關係是以多項式表示。此外,基於電池的非線性時變特性,我們使用擴展卡爾曼濾波器(extended kalman filtering, EKF)估測電量狀態,並將估測結果與實際量測數據比較,驗證所提出的方法正確可行。
This thesis proposes a method for estimating state of charge (SOC) of lithium-ion battery. Equivalent circuit model is used to represent the dynamic behavior of the electro-chemical reaction of the lithium ion batteries. Parameters of the equivalent circuit are obtained by least square fit of voltage, current data gathered from battery charging and discharging test platform. The relation between those parameters (eg. Voc, R and C’s) to identified and independent variables, such as SOC and charging / discharging rates, are represented by polynomials. Due to the nonlinear and time-varying characteristics of the battery cell, extended Kalman Filter (KF) are employed to estimate the SOC. Comparison results between the estimated SOC and data from experiments shows that the proposed SOC estimation method is feasible and accurate.