A 1D inverse finite-element framework for shape-sensing and assessment of multiple degrading boundary conditions in multi-span beam-like structures
Structural Health Monitoring · 20.11.2026
DOI: 10.1177/14759217261500630 ↗Kurz gefasst
Fachartikel in der Zeitschrift Structural Health Monitoring, online erschienen am 20.11.2026. Verfasst von Hung Chih Wu Chen und 3 weiteren Autor:innen. Die Arbeit nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
In multi-span beam-like structures, continuity couples the response of adjacent spans; hence, degradation or settlement at one or more supports can alter load-transfer throughout the structure. Since strain-based shape-sensing methods commonly assume ideal support constraints, these changes can bias the reconstructed deformation field, although the actual condition of the supports remains unknown. This study develops a one-dimensional inverse finite-element method (1D iFEM) framework for simultaneous deformation reconstruction and stiffness-based condition assessment of multiple predefined candidate vertical supports. The associated support constraints are released so that their settlements are reconstructed directly from strain measurements. When the resulting displacement field is not uniquely defined, independent displacement measurements are introduced as kinematic anchors, with one anchor for each remaining strain-unobservable displacement mode. Support reactions are recovered from the reconstructed curvature and were combined with the estimated settlements to determine the equivalent vertical stiffnesses. The framework is experimentally validated on a two-span aluminium beam instrumented with fibre Bragg grating strain sensors, with individual and combined-support-compliance configurations representative of degradation reproduced by vertical springs under static and quasi-static moving loading. Under repeated static loading, the mean normalized mean absolute displacement error ( NMAE v ) ranged from 0 . 96 % to 4 . 94 % , the maximum mean local laser-validation error was 6 . 84 % and the maximum relative difference between the mean estimated and target support stiffnesses was 5 . 63 % . Under moving loading, displacement root-mean-square errors of 0 . 20 mm and 0 . 25 mm were obtained at two independent validation locations, whereas the median stiffness differences were 1 . 26 % and 6 . 27 % for engaged and sufficiently excited supports. Monte Carlo simulations quantify sensitivity to strain noise and displacement–anchor uncertainty, and the workflow requires approximately 13 ms per update. The framework therefore enables deformation reconstruction under non-ideal boundary conditions and equivalent stiffness-based assessment of multiple degrading supports.
Zitieren
APA
Chen, H. C. W., Bardiani, J., Mastrorillo, L., & Sbarufatti, C. (2026). A 1D inverse finite-element framework for shape-sensing and assessment of multiple degrading boundary conditions in multi-span beam-like structures. Structural Health Monitoring. https://doi.org/10.1177/14759217261500630
Vancouver
Chen HCW, Bardiani J, Mastrorillo L, Sbarufatti C. A 1D inverse finite-element framework for shape-sensing and assessment of multiple degrading boundary conditions in multi-span beam-like structures. Structural Health Monitoring. 2026. doi:10.1177/14759217261500630
BibTeX
@article{Chen2026,
author = {Chen, Hung Chih Wu and Bardiani, Jacopo and Mastrorillo, Luca and Sbarufatti, Claudio},
title = {A 1D inverse finite-element framework for shape-sensing and assessment of multiple degrading boundary conditions in multi-span beam-like structures},
journal = {Structural Health Monitoring},
year = {2026},
doi = {10.1177/14759217261500630}
}
EUVIMED