| Title |
SOH Estimation of LFP Battery Packs for Electric Vehicles Using a CNN-LSTM Model |
| Authors |
Hee-Sung Lim ; Kyo-Beom Lee |
| DOI |
http://doi.org/10.5207/JIEIE.2025.39.4.266 |
| Keywords |
AI; Battery; CNN; LFP; LSTM; SOC; SOH |
| Abstract |
This paper proposes a method for estimating the SOH (State of Health) of an LFP battery pack for electric vehicles using AI algorithms. Among the AI algorithms, CNN receives the battery's time series data as an image and extracts features to estimate the battery's state. LSTM(Long Short-Term Memory) can estimate electrochemical parameters in real-time based on time series data and learns the output time series data of CNN and the battery's charge/discharge cycle data to predict the remaining life of the battery. As a result of the estimation, we confirmed that when estimating SOH using the LSTM model, higher SOH estimation accuracy can be obtained compared to linear regression analysis. |