The Battery State of Charge Estimation Based Weighted Least Squares Support Vector Machine
Authors:
- Yongqiang ChenYCYongqiang Chen
- Bo LongBLBo Long
- Xiao LeiXLXiao Lei
- 1 March 2011
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21574839,p. 1-4
Abstract
A new method to estimate the battery state of charge (SOC) in electric vehicles (EV) based on support vector machine is presented. The key of the proposed method is to establish the relationship of the SOC to the battery current, voltage and temperature by using weighted least squares support vector machine (WLS-SVM). With the goal of achieving the optimal robust estimation of the SOC, the extended Huber estimation of residual is employed instead of sum of the least square of the residual in the objective function of LS-SVM. And the iterative modeling algorithm is proposed. The result shows that the proposed estimator can stimulate the battery dynamics for the accurate estimation of SOC in EV.Keywords
This publication has 10 references indexed in Scilit:
- Support Vector Machines: Theory and Applications2005
- State-of-charge and state-of-health prediction of lead-acid batteries for hybrid electric vehicles using non-linear observers2005
- Extended Kalman filtering for battery management systems of LiPB-based HEV battery packsJournal of Power Sources, 2004
- Kernel Methods for Pattern Analysis2004
- Impedance-based non-linear dynamic battery modeling for automotive applicationsJournal of Power Sources, 2003
- A battery ageing model used in stand alone PV systemsJournal of Power Sources, 2002
- Weighted least squares support vector machines: robustness and sparse approximationNeurocomputing, 2002
- The state of the art of electric and hybrid vehiclesProceedings of the IEEE, 2002
- Modern Electric Vehicle Technology2001
- Energy gauge for lead-acid batteries in electric vehiclesIEEE Transactions on Energy Conversion, 2000