In order to solve the problems that lithium-ion power battery cannot reflect state of health(SOH) in sorting process , the parameters which can reflect the battery SOH, such as capacity, AC resistance and self-discharge current, were used as the input vector of battery sorting model, fuzzy c-means clustering analysis and support vector machine based on cross validation algorithm to the battery for classification and recognition were used, and lithium power battery sorting model was established and the same batch of power battery were separating tested and according to the experimental results, batteries was divided into groups. The test results showed that: battery electrochemistry was having a good consistency. The variation of capacitance was less than 5% while there was 1500 cycle life.
[1]
Sun Zechang,et al.
Two Methods Studying Internal Resistance for Lithium-ion Battery Life Estimation Used in EV
,
2011
.
[2]
Ralph E. White,et al.
Mathematical modeling of the capacity fade of Li-ion cells
,
2003
.
[3]
Haiying Wang,et al.
Lithium Battery SOC Estimation Method Study Based on Principal Component Analysis
,
2015
.
[4]
Haiying Wang,et al.
Estimation of State of Charge of Batteries for Electric Vehicles
,
2013
.
[5]
Dae-Keun Kang,et al.
Investigation on cell impedance for high-power lithium-ion batteries
,
2007
.