The State of Charge Estimation of Li-Ion Batteries in Electric Vehicle Based on RBF Neural Network Optimized by Genetic Algorithm

Jun Bi, Qiuping Xu, Kai Wang, Dong ZhangInternational Journal of Digital Content Technology and its Applications 2013 · DOI

Abstract

In order to guarantee security and stable operation of electric vehicle, it is necessary to on-line estimation for the state of charge (SOC) of batteries. The power battery is a complex nonlinear system, and Radial Basis Function Neural Network (RBF NN) has advantages in solving nonlinear problems, so the model of on-line SOC estimation based on RBF NN is proposed. In order to improve the prediction accuracy of SOC, genetic algorithm is used to optimize the model, which can make global optimization search for the center and spread of each neuron in hidden layer of the RBF NN to get the most optimal value. The experiments are based on the battery data achieved from the pure electric buses with LiFePO4 Li-ion batteries running during the period of 2010 Shanghai World Expo. The results show that compared with RBF NN, RBF NN optimized by genetic algorithm significantly can improve the prediction accuracy of SOC.
BibTeX
@article{bi2013state,
  author = {Jun Bi and Qiuping Xu and Kai Wang and Dong Zhang},
  title = {{The State of Charge Estimation of Li-Ion Batteries in Electric Vehicle Based on RBF Neural Network Optimized by Genetic Algorithm}},
  journal = {International Journal of Digital Content Technology and its Applications},
  year = {2013},
  doi = {10.4156/jdcta.vol7.issue5.44},
}

Loading related content...

Figure