Physics-Informed Machine Learning for Structural Health Monitoring

Cross, Elizabeth; Gibson, Samuel J.; Jones, Matthew; Pitchforth, Daniel J.; Zhang, S.; Rogers, Timothy J.

doi:10.1007/978-3-030-81716-9_17

Cited by 25 publications

(21 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, going beyond the detection stage with unsupervised techniques is still a challenge, and damage prognosis is still achieved only when the physics of damage progression is included in a hybrid data/model-driven approach. Physics-informed ML applications are working towards this goal [21].…”

Section: In the Shm Processmentioning

confidence: 99%

A proposal of classification for machine-learning vibration-based damage identification methods

Marafini¹

2023

Materials Research Proceedings

View full text Add to dashboard Cite

Abstract. Recent advances in computing power and sensing technology led to a significant evolution of Structural Health Monitoring (SHM) techniques, transforming SHM into a “Big Data” problem. The use of data-driven approaches for damage identification purposes, specifically Machine Learning (ML) methods, has gained popularity. ML can help at various levels of the SHM process: to pre- and post-process input data, extract damage sensitive features, and operate pattern recognition in measured data and output valuable information for damage identification. In this paper, the role of ML in SHM applications is discussed together with a new scheme for classifying ML applications in SHM, especially focusing on vibration-based monitoring, given its consolidated theoretical base. Finally, the implications of the application of these methods to historic structures are discussed, with a brief account of existing case studies. The proposed classification is exemplified using the most recent studies available in the literature on cultural heritage structures.

show abstract

Section: In the Shm Processmentioning

confidence: 99%

A proposal of classification for machine-learning vibration-based damage identification methods

Marafini¹

2023

Materials Research Proceedings

View full text Add to dashboard Cite

show abstract

“…SHM techniques aim to infer the state of health of a structure given measured data from sensors; such techniques may be clustered in data-based black-box models, white-box or physics-based models, and mixed data and physics-based models, also known as grey-box models [ 13 , 14 ]. Black-box methodologies generally make use of Machine Learning techniques to extract damage-sensitive features from the signals’ pattern, with no information on the underlying physics of the problem at hand.…”

Section: Introductionmentioning

confidence: 99%

“…In contrast, white-box models make use of physics to interpret signal patterns and infer the health state of structures. Grey-box models combine purely data-based techniques with physics-based ones [ 13 , 14 ].…”

Section: Introductionmentioning

confidence: 99%

Variable Thickness Strain Pre-Extrapolation for the Inverse Finite Element Method

Poloni

Oboe

Sbarufatti

et al. 2023

Sensors

View full text Add to dashboard Cite

The inverse Finite Element Method (iFEM) has recently gained much popularity within the Structural Health Monitoring (SHM) field since, given sparse strain measurements, it reconstructs the displacement field of any beam or shell structure independently of the external loading conditions and of the material properties. However, in principle, the iFEM requires a triaxial strain measurement for each inverse finite element, which is seldom feasible in practical applications due to both costs and cabling-related limitations. To alleviate this problem several techniques to pre-extrapolate the measured strains have been developed, so that interpolated or extrapolated strain values are inputted to elements without physical sensors: the benefit is that the required number of sensors can be reduced. Nevertheless, whenever the monitored components comprise regions of different thicknesses, each region of constant thickness must be extrapolated separately, due to thickness-induced discontinuities in the strain field. This is the case in many practical applications, especially those concerning fiber-reinforced composite laminates. This paper proposes to extrapolate the measured strain field in a thickness-normalized space, where the thickness-induced trends are removed; this novel method can significantly decrease the number of required sensors, effectively reducing the costs of iFEM-based SHM systems. The method is validated in a simple but informative numerical case study, highlighting the potentialities and benefits of the proposed approach for more complex application scenarios.

show abstract

“…The research in this category traditionally adopts modern signal processing methods and tries to catch a sudden change in signals caused by the occurrence of damage [ 35 , 36 ]. Recently, methodologies based on data analysis and information extraction, in the broad field of machine learning, are being increasingly used to address damage/failure identification problems to achieve a wider range of applicability [ 37 , 38 ]. In order to overcome the limitations associated to traditional neural networks solutions [ 39 ], such as real-world noise, more complex deep learning (DL) models and techniques, with higher generalisation capabilities, have been introduced as data extractors, classifiers, and predictors [ 40 , 41 , 42 ].…”

Section: Introductionmentioning

confidence: 99%

A Study on Structural Health Monitoring of a Large Space Antenna via Distributed Sensors and Deep Learning

Angeletti

Iannelli

Gasbarri

et al. 2022

Sensors

View full text Add to dashboard Cite

Most modern Earth and Universe observation spacecraft are now equipped with large lightweight and flexible structures, such as antennas, telescopes, and extendable elements. The trend of hosting more complex and bigger appendages, essential for high-precision scientific applications, made orbiting satellites more susceptible to performance loss or degradation due to structural damages. In this scenario, Structural Health Monitoring strategies can be used to evaluate the health status of satellite substructures. However, in particular when analysing large appendages, traditional approaches may not be sufficient to identify local damages, as they will generally induce less observable changes in the system dynamics yet cause a relevant loss of payload data and information. This paper proposes a deep neural network to detect failures and investigate sensor sensitivity to damage classification for an orbiting satellite hosting a distributed network of accelerometers on a large mesh reflector antenna. The sensors-acquired time series are generated by using a fully coupled 3D simulator of the in-orbit attitude behaviour of a flexible satellite, whose appendages are modelled by using finite element techniques. The machine learning architecture is then trained and tested by using the sensors’ responses gathered in a composite scenario, including not only the complete failure of a structural element (structural break) but also an intermediate level of structural damage. The proposed deep learning framework and sensors configuration proved to accurately detect failures in the most critical area or the structure while opening new investigation possibilities regarding geometrical properties and sensor distribution.

show abstract

Physics-Informed Machine Learning for Structural Health Monitoring

Cited by 25 publications

References 57 publications

A proposal of classification for machine-learning vibration-based damage identification methods

A proposal of classification for machine-learning vibration-based damage identification methods

Variable Thickness Strain Pre-Extrapolation for the Inverse Finite Element Method

A Study on Structural Health Monitoring of a Large Space Antenna via Distributed Sensors and Deep Learning

Contact Info

Product

Resources

About