Proceedings of the KIEE Conference (대한전기학회:학술대회논문집)

The Korean Institute of Electrical Engineers (대한전기학회)
권호 표지

State of Charge Indicator for Electric Vehicle using Neural Networks

신경회로망을 이용한 전기자동차용 바테리 잔존용량계

  • Byun, Sung-Chun (Dept. of Electrical Engineering, Chonnam National University) ;
  • Kim, Eui-Sun (Dept. of Electrical Engineering, Chonnam National University) ;
  • Ryoo, Young-Jae (Dept. of Electrical Engineering, Chonnam National University) ;
  • Lim, Young-Cheol (Dept. of Electrical Engineering, Chonnam National University)
  • Published : 1998.07.20
Download PDF
⟨ Previous Next ⟩

Abstract

A new approach to developing battery SOC indicator for electric vehicle is discussed in this paper. One of the most difficult problems associated with the development of electric vehicle is the battery indicator which reliably informs the state of charge(SOC) of the battery to the driver. And the condition to be satisfied with SOC indicator installed on the electric vehicle is that it should be used under frequently variable load. A new method to determining SOC using neural networks(NN) is proposed to satify the condition. The training data of NN are obtained by using mathematical model of lead-acid battery, and calculating discharge currents and terminal voltages while battery discharges with constant current. The 3-layered NN with back propagation algorithm is used Simulation results show that the proposed method is appropriate as SOC indicator of the battery.

Keywords