From LLM to FEM: Low-Rank Adaptation for Noise-Robust Structural Damage Detection
Sensors · 11.03.2026 · 26 (6) · 1776
DOI: 10.3390/s26061776 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 11.03.2026. Verfasst von Jaedong Kim und 2 weiteren Autor:innen. Beteiligte Einrichtung: Energy & Environment Laboratory, KEPCO Research Institute, Daejeon 34056, Republic of Korea. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Korea Institute of Energy Technology Evaluation and Planning.
Abstract (Originalsprache)
Structural damage detection using the finite element method is inherently formulated as an inverse problem, often suffering from ill-posedness and high sensitivity to measurement noise. This study introduces a novel damage detection methodology by applying low-rank adaptation (LoRA), originally developed for fine-tuning large language models, to inverse problems in structural mechanics for the first time. The proposed approach exploits the physically inherent low-rank nature of structural damage: damage is typically localized, and the contribution of each finite element to the stiffness matrix is limited by its degrees of freedom. Accordingly, the stiffness change matrix is factorized into two low-rank matrices, reducing the number of parameters and providing implicit regularization against full-rank measurement noise. Physical consistency is ensured through sparsity and symmetry constraints. Numerical experiments on cantilever beam and L-shaped plate structures across five damage scenarios demonstrated that the proposed method achieved superior noise robustness compared with baseline methods. At a signal-to-noise ratio of 20 dB, representative of practical field conditions, LoRA achieved stiffness errors below 2%, whereas the baseline methods failed to provide reliable results. The proposed framework achieved a 100% success rate in damage zone localization (Precision@n ≥ 80%) with over 60% parameter reduction, presenting a promising solution for practical structural health monitoring.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Energy & Environment Laboratory, KEPCO Research Institute, Daejeon 34056, Republic of Korea.
- R&D Strategy Center, KEPCO Research Institute, Daejeon 34056, Republic of Korea.
ORCID-Profile: Jaedong Kim, Haesu Kang
Förderung
- Korea Institute of Energy Technology Evaluation and Planning
Zitieren
APA
Kim, J., Kang, H., & Chang, S. (2026). From LLM to FEM: Low-Rank Adaptation for Noise-Robust Structural Damage Detection. Sensors, 26(6), 1776. https://doi.org/10.3390/s26061776
Vancouver
Kim J, Kang H, Chang S. From LLM to FEM: Low-Rank Adaptation for Noise-Robust Structural Damage Detection. Sensors. 2026;26(6):1776. doi:10.3390/s26061776
BibTeX
@article{Kim2026,
author = {Kim, Jaedong and Kang, Haesu and Chang, Sungyong},
title = {From LLM to FEM: Low-Rank Adaptation for Noise-Robust Structural Damage Detection},
journal = {Sensors},
year = {2026},
volume = {26},
number = {6},
pages = {1776},
doi = {10.3390/s26061776}
}
EUVIMED