A population-based SHM methodology for heterogeneous structures: Transferring damage localisation knowledge between different aircraft wings

Gardner, Paul A.; Bull, Larry; Gosliga, Julian; Poole, Jack; Dervilis, Nikolaos; Worden, Keith

doi:10.1016/j.ymssp.2022.108918

Cited by 26 publications

(15 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Domain adaptation was applied in PBSHM to localize damage across a heterogeneous population of aircraft wings in Gardner et al (2022). The authors employed graph matching methods alongside domain adaptation techniques (specifically, balanced distribution adaptation; Wang et al, 2017) in order to identify the most suitable location labels to transfer information across in an unsupervised manner, that is, without using target damage labels.…”

Section: Related Workmentioning

confidence: 99%

“…Typically, statistical properties of the data in the source and target are compared (Pan et al, 2010), which can be extended to also comparing the similarities in the geometric structure of the data (Long et al, 2013). For monitoring applications, the majority of work on transfer learning has focused on fine-tuning of neural networks (Cao et al, 2018; Dorafshan et al, 2018; Gao and Mosalam, 2018; Zhu et al, 2020) and domain adaptation (Michau and Fink, 2019; Gardner et al, 2020; Bull et al, 2021; Xu and Noh, 2021; Gardner et al, 2022). Since this article aims to leverage labeled source data where there are no labeled target data, domain adaptation is the most appropriate technology to investigate.…”

Section: Introductionmentioning

confidence: 99%

“…Domain adaptation, which is a branch of techniques within transfer learning, has gained popularity as a useful method to reduce the difference between a labeled source and an unlabelled target domain, in order to transfer information across. It is used abundantly in visual applications (Ganin et al, 2016; Long et al, 2016; Csurka, 2017) and natural language processing (Ben-David et al, 2006; Blitzer et al, 2007), fault detection/condition monitoring (Li et al, 2019; Jiao et al, 2020; Li et al, 2020; Wang and Liu, 2020; Ding et al, 2021; Zhang and Li, 2022; Li et al, 2023) and PBSHM (Michau and Fink, 2019; Gardner et al, 2020; Bull et al, 2021; Xu and Noh, 2021; Gardner et al, 2022). In PBSHM, domain adaptation—in the form of transfer component analysis—was used to detect damage on a tailplane with incomplete data, by leveraging information from a heterogeneous population of other tailplanes (Bull et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Damage localisation using disparate damage states via domain adaptation

Wickramarachchi,

Gardner,

Poole

et al. 2024

DCE

View full text Add to dashboard Cite

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.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Damage localisation using disparate damage states via domain adaptation

Wickramarachchi,

Gardner,

Poole

et al. 2024

DCE

View full text Add to dashboard Cite

show abstract

“…Domain adaptation has also been demonstrated in a PBSHM setting, Gardner et al have shown that DA can be used to transfer localisation labels between numerical and experimental structures, 30 two heterogeneous aircraft wings 31 and between pre- and post-repair states in aircraft wings. 32 In Bull et al , a population of six experimental tailplanes was used to demonstrate transferring a damage detector.…”

Section: Introductionmentioning

confidence: 99%

On statistic alignment for domain adaptation in structural health monitoring

Poole

Gardner

Dervilis

et al. 2022

Structural Health Monitoring

View full text Add to dashboard Cite

The practical application of structural health monitoring is often limited by the availability of labelled data. Transfer learning – specifically in the form of domain adaptation (DA) – gives rise to the possibility of leveraging information from a population of physical or numerical structures, by inferring a mapping that aligns the feature spaces. Typical DA methods rely on nonparametric distance metrics, which require sufficient data to perform density estimation. In addition, these methods can be prone to performance degradation under class imbalance. To address these issues, statistic alignment (SA) is discussed, with a demonstration of how these methods can be made robust to class imbalance, including a special case of class imbalance called a partial DA scenario. Statistic alignment is demonstrated to facilitate damage localisation with no target labels in a numerical case study, outperforming other state-of-the-art DA methods. It is then shown to be capable of aligning the feature spaces of a real heterogeneous population, the Z24 and KW51 bridges, with only 220 samples used from the KW51 Bridge. Finally, in scenarios where more complex mappings are required for knowledge transfer, SA is shown to be a vital pre-processing tool, increasing the performance of established DA methods.

show abstract

“…DA has also been demonstrated in a PBSHM setting, Gardner et al have shown that DA can be used to transfer localisation labels between numerical and experimental structures [30], two heterogeneous aircraft wings [31], and between preand post-repair states in aircraft wings [32]. In Bull et al a population of six experimental tailplanes was used to demonstrate transferring a damage detector [33].…”

Section: Introductionmentioning

confidence: 99%

On statistic alignment for domain adaptation in structural health monitoring

Poole¹,

Gardner²,

Dervilis³

et al. 2022

Preprint

View full text Add to dashboard Cite

The practical application of structural health monitoring (SHM) is often limited by the availability of labelled data. Transfer learning -specifically in the form of domain adaptation (DA) -gives rise to the possibility of leveraging information from a population of physical or numerical structures, by inferring a mapping that aligns the feature spaces. Typical DA methods rely on nonparametric distance metrics, which require sufficient data to perform density estimation. In addition, these methods can be prone to performance degradation under class imbalance. To address these issues, statistic alignment (SA) is discussed, with a demonstration of how these methods can be made robust to class imbalance, including a special case of class imbalance called a partial DA scenario. SA is demonstrated to facilitate damage localisation with no target labels in a numerical case study, outperforming other state-of-the-art DA methods. It is then shown to be capable of aligning the feature spaces of a real heterogeneous population, the Z24 and KW51 bridges, with only 220 samples used from the KW51 bridge. Finally, in scenarios where more complex mappings are required for knowledge transfer, SA is shown to be a vital pre-processing tool, increasing the performance of established DA methods.

show abstract

A population-based SHM methodology for heterogeneous structures: Transferring damage localisation knowledge between different aircraft wings

Cited by 26 publications

References 41 publications

Damage localisation using disparate damage states via domain adaptation

Damage localisation using disparate damage states via domain adaptation

On statistic alignment for domain adaptation in structural health monitoring

On statistic alignment for domain adaptation in structural health monitoring

Contact Info

Product

Resources

About