Details
| Original language | English |
|---|---|
| Article number | e3 |
| Journal | Data-Centric Engineering |
| Volume | 5 |
| Issue number | 1 |
| Publication status | Published - 2 Feb 2024 |
Abstract
A significant challenge of structural health monitoring (SHM) is the lack of labeled data collected from damage states. Consequently, the collected data can be incomplete, making it difficult to undertake machine learning tasks, to detect or predict the full range of damage states a structure may experience. Transfer learning is a helpful solution, where data from (source) structures containing damage labels can be used to transfer knowledge to (target) structures, for which damage labels do not exist. Machine learning models are then developed that generalize to the target structure. In practical applications, it is unlikely that the source and the target structures contain the same damage states or experience the same environmental and operational conditions, which can significantly impact the collected data. This is the first study to explore the possibility of transfer learning for damage localisation in SHM when the damage states and the environmental variations in the source and target datasets are disparate. Specifically, using several domain adaptation methods, this article localizes severe damage states at a target structure, using labeled information from minor damage states at a source structure. By minimizing the distance between the marginal and conditional distributions between the source and the target structures, this article successfully localizes damage states of disparate severities, under varying environmental and operational conditions. The effect of partial and universal domain adaptation - where the number of damage states in the source and target datasets differ - is also explored in order to mimic realistic industrial applications of these methods.
Keywords
- PBSHM, SHM, damage localisation, distance metrics, domain adaptation
ASJC Scopus subject areas
- Engineering(all)
- Mathematics(all)
- Applied Mathematics
- Mathematics(all)
- Statistics and Probability
- Computer Science(all)
- Computer Science Applications
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Data-Centric Engineering, Vol. 5, No. 1, e3, 02.02.2024.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Damage localisation using disparate damage states via domain adaptation
AU - Wickramarachchi, Chandula t.
AU - Gardner, Paul
AU - Poole, Jack
AU - Hübler, Clemens
AU - Jonscher, Clemens
AU - Rolfes, Raimund
N1 - Funding Information: We gratefully acknowledge the financial support of the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)—SFB-1463—434502799. C.T.W. would also like to thank the Mercator fellowship program as well as The Dynamics Research Group at the University of Sheffield for supporting this work.
PY - 2024/2/2
Y1 - 2024/2/2
N2 - A significant challenge of structural health monitoring (SHM) is the lack of labeled data collected from damage states. Consequently, the collected data can be incomplete, making it difficult to undertake machine learning tasks, to detect or predict the full range of damage states a structure may experience. Transfer learning is a helpful solution, where data from (source) structures containing damage labels can be used to transfer knowledge to (target) structures, for which damage labels do not exist. Machine learning models are then developed that generalize to the target structure. In practical applications, it is unlikely that the source and the target structures contain the same damage states or experience the same environmental and operational conditions, which can significantly impact the collected data. This is the first study to explore the possibility of transfer learning for damage localisation in SHM when the damage states and the environmental variations in the source and target datasets are disparate. Specifically, using several domain adaptation methods, this article localizes severe damage states at a target structure, using labeled information from minor damage states at a source structure. By minimizing the distance between the marginal and conditional distributions between the source and the target structures, this article successfully localizes damage states of disparate severities, under varying environmental and operational conditions. The effect of partial and universal domain adaptation - where the number of damage states in the source and target datasets differ - is also explored in order to mimic realistic industrial applications of these methods.
AB - A significant challenge of structural health monitoring (SHM) is the lack of labeled data collected from damage states. Consequently, the collected data can be incomplete, making it difficult to undertake machine learning tasks, to detect or predict the full range of damage states a structure may experience. Transfer learning is a helpful solution, where data from (source) structures containing damage labels can be used to transfer knowledge to (target) structures, for which damage labels do not exist. Machine learning models are then developed that generalize to the target structure. In practical applications, it is unlikely that the source and the target structures contain the same damage states or experience the same environmental and operational conditions, which can significantly impact the collected data. This is the first study to explore the possibility of transfer learning for damage localisation in SHM when the damage states and the environmental variations in the source and target datasets are disparate. Specifically, using several domain adaptation methods, this article localizes severe damage states at a target structure, using labeled information from minor damage states at a source structure. By minimizing the distance between the marginal and conditional distributions between the source and the target structures, this article successfully localizes damage states of disparate severities, under varying environmental and operational conditions. The effect of partial and universal domain adaptation - where the number of damage states in the source and target datasets differ - is also explored in order to mimic realistic industrial applications of these methods.
KW - PBSHM
KW - SHM
KW - damage localisation
KW - distance metrics
KW - domain adaptation
UR - http://www.scopus.com/inward/record.url?scp=85185222657&partnerID=8YFLogxK
U2 - 10.1017/dce.2023.29
DO - 10.1017/dce.2023.29
M3 - Article
VL - 5
JO - Data-Centric Engineering
JF - Data-Centric Engineering
IS - 1
M1 - e3
ER -