Abnormal Degradation Detection for Lithium-Ion Batteries With Denoising Diffusion Implicit Model
- Authors
- Jang, Kyujin; Park, Jongwook; Bae, Sungwoo
- Issue Date
- Jan-2026
- Publisher
- Institute of Electrical and Electronics Engineers Inc.
- Keywords
- anomaly detection; denoising diffusion implicit model; lithium-ion battery; time-series anomaly detection
- Citation
- 2025 28th International Conference on Electrical Machines and Systems (ICEMS), pp 3321 - 3324
- Pages
- 4
- Indexed
- SCOPUS
- Journal Title
- 2025 28th International Conference on Electrical Machines and Systems (ICEMS)
- Start Page
- 3321
- End Page
- 3324
- URI
- https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/211831
- DOI
- 10.23919/ICEMS66262.2025.11317751
- ISSN
- 2640-7841
2642-5513
- Abstract
- This paper proposes a method for the immediate detection of abnormal degradation in lithium-ion batteries by applying the denoising diffusion implicit model (DDIM). The proposed approach utilizes a diffusion-based generative model to emphasize and reconstruct battery operation data. In this framework, the DDIM compresses and reconstructs lithium-ion battery operational data, and the resulting reconstruction error is used as an indicator of anomalies. Experimental results demonstrate that the proposed method can detect abnormal degradation at least 10 cycles earlier than the benchmark models.
- Files in This Item
-
Go to Link
- Appears in
Collections - 서울 공과대학 > 서울 전기공학전공 > 1. Journal Articles

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.