| 제목 |
리튬 폴리머 배터리의 SOC 추정을 위한 확장 칼만 필터의 LSTM 기반 보정방법 |
| 분야 |
전기동력자동차 |
| 언어 |
Korean |
| 저자 |
윤범진(한국기술교육대학교), 유승열(한국기술교육대학교), 성상만(한국기술교육대학교) |
| Key Words |
Electric vehicle(전기자동차), Neural network(인공신경망), Long short-term memory(장단기 기억), Extended kalman filter(확장칼만필터), SOC(충전량) |
| 초록 |
State of charge(SOC) and open circuit voltage(OCV) are essential for maintaining the performance and preventing battery malfunction in an electric vehicle(EV). Estimation methods about battery have been studied for decades; however, it is still difficult to estimate precise SOC immediately after the discharge. The extended Kalman filter(EKF) is generally used for diagnosis, but its accuracy can be reduced based on the parameter of the battery equivalent model. EKF, which is compensated by the long short-term memory(LSTM) network, is proposed in this paper. LSTM trains the tendency of the battery’s state and proceeds in a way that corrects the result of the EKF according to the load profile after discharging. This proposed method is verified through a comparison with the actual OCV and SOC. The findings of this experiment show the decrease in the margin of error when compared with normal EKF based on different load profiles. |
| 원문(PDF) |
다운로드
Journal Site
|
