| Title |
MMF Model : Multi-Modal Fusion Model for Lithium-Ion Battery SOH Estimation |
| Authors |
김차니(Chani Kim) ; 설수진(Sujin Seol) ; 김병우(Byeong-Woo Kim) |
| DOI |
https://doi.org/10.5370/KIEE.2025.74.11.1926 |
| Keywords |
Lithium-ion Battery; SOH estimation; Deep learning; Multi-modal |
| Abstract |
With the increasing demand for lithium-ion batteries, accurate estimation of battery SOH has become essential for ensuring operational safety and system reliability. However, SOH prediction based on a single type of sensor data often suffers from limited accuracy and insufficient robustness. To address this limitation, it is necessary to incorporate diverse sensor information that reflects the comprehensive state of the battery. In this study, we propose a Multi-modal Fusion (MMF) model that effectively leverages multiple types of sensor data related to battery behavior. The proposed model integrates heterogeneous modalities in a complementary manner, and experimental results demonstrate that this MMF model approach significantly improves prediction accuracy compared to models using individual data types. |