Population-Based Structural Health Monitoring Research Articles

Transfer learning in bridge monitoring: Laboratory study on domain adaptation for population-based SHM of multispan continuous girder bridges

The presence of sufficient labelled data associated to various environmental conditions and damage scenarios often represents a challenge for the applicability of supervised-learning methods when dealing with structural health monitoring of real-scale infrastructures. To address this problem, population-based structural health monitoring has been recently proposed as an attractive solution, with the goal to collect information from similar structures and transfer health-state labels across the population. This paper focusses on the use of a feature-based transfer learning method. A machine-learning model is trained with source labelled data in a transformed features space to afterwards classify the unlabelled target dataset of a different bridge. More specifically, a domain adaptation-based methodology proposing two possible strategies, a single-source or a multi-source approach, is described. Given the difficulties in validating these techniques on real and varied datasets from multiple bridges, this paper presents a physical benchmark for population-based structural health monitoring in civil-engineering applications. The transfer between different configurations of a laboratory-scale bridge model, subjected to multiple experimental tests under changing environmental conditions and to the same pseudo-damage scenarios, is investigated. The results of the experimental campaign demonstrate the possibility of effectively exchanging damage labels to perform novelty detection and damage classification across the population via domain adaptation, improve the identification of specific damage classes, as well as to increase model’s outcome using a multi-source approach, thus overcoming the limitations of conventional machine learning-based methods. Furthermore, this paper provides an open dataset with the physical benchmark-related data, allowing other researchers to test their own algorithms and address the various transfer learning challenges.

Exploring the effective implementation of population-based SHM in existing buildings. Part I: Structural condition assessment

Despite major critical progress over the past few decades, condition-based decision-making concerning the immediate occupancy or efficient operation of an existing building continues to face unresolved issues. The key issue is to determine the structural condition and its residual capacity using data-driven-based methods without knowledge of its life cycle since design (ageing, extreme events, etc.). Environmental and operational variations, boundary conditions and a lack of damage-state building data lead to further uncertainties in the actual assessment of the structural capacity and affect the effectiveness of implementing structural health monitoring (SHM) solution. Recently, population-based SHM has been developed to address some of these issues to enable transfer from data collected from a group of nominally identical buildings, and to model missing data for the target building. This study presents the Build’Health™ framework for operational data-driven SHM, considering populations of buildings evaluated under operational conditions. Fifty-five period-height empirical models obtained from methods based on ambient vibrations recorded in nominally identical reinforced concrete buildings were first identified in the literature to define the building population models, along with design features (such as boundary conditions, geometry or material design). The shift in structural condition of the target building is inferred from the nominally identical building population, after adding empirical variability in the resonance period due to temperature, assumed herein to be the main environmental cause of structural variation. This work presents the framework implemented and its application to five target buildings under different monitoring and structural conditions. This manuscript is the first of a two-part article on population-based SHM relative to structural condition assessment (part I) and damage-feature classification (part II) presenting the Build’Health™ concept.

Similarity assessment of structures for population-based structural health monitoring via graph kernels

Similarity assessment of structures for population-based structural health monitoring via graph kernels

Assessing structural homogeneity and heterogeneity in offshore wind farms: A population-based structural health monitoring approach

This paper presents an in-depth analysis of population-based structural health monitoring applied to offshore wind farms, with a specific focus on a real wind farm. The study quantifies the heterogeneity within the wind turbine population utilising advanced numerical methods. The Fréchet distance is introduced as a key metric to assess the similarity in terms of structural integrity among the turbines considering factors, such as geometry, material properties, topology, and operational conditions. The results revealed a predominant homogeneity with respect across the wind turbines, despite variations in environmental exposure and operational states. Furthermore, the scope of the study is extended to explore the implications of environmental influences, such as corrosion, scour, and marine growth, which are typically unmonitored yet crucial for comprehensive structural assessment. The present paper contributes to the field by demonstrating the feasibility of knowledge transfer in homogeneous wind turbine populations and highlighting the need for detailed similarity assessments in heterogeneous settings. The findings of this study pave the way for better-targeted, smart, and efficient structural health monitoring strategies in renewable energy systems.

Incorporating specialist engineering knowledge into IE models to facilitate enhanced SHM-based transfer learning within PBSHM

Population-based Structural Health Monitoring (PBSHM) aims to gather additional knowledge on a structure's health by monitoring multiple structures (the population), compared to the level of knowledge available when utilising only a single structure's data. Before effective transfer learning can occur, a similarity between the structures must be established to prevent negative transfer. Irreducible Element (IE) models, combined with graph theory, are the vehicle used within PBSHM to facilitate this process. Recent research has introduced the Canonical Form (CF) which facilitates a singular IE model per structure; however, detailed IE models solely rely on predefined rules to reduce down to the CF, and do not require the incorporation of specialist engineering knowledge within these reductions. In particular, the specific SHM problem of interest may constrain the reduction. This work aims to enhance the capabilities of IE models by allowing authors to incorporate their specialist engineering knowledge into an IE model. A case study of a steel truss pedestrian bridge is presented, focusing on incorporating the engineering knowledge utilised in its design, construction, and maintenance within an IE model. By enabling authors to leverage their domain expertise, a more comprehensive understanding of structural similarities can be achieved, thereby enhancing transfer learning potential in PBSHM. This paper outlines the methodology and provides insights into the application of domain knowledge in IE models, showcasing its potential benefits for the field.

Measuring data similarity in population-based structural health monitoring using distance metrics

Population-based structural health monitoring (PBSHM) expands structural health monitoring (SHM) from a single structure to a group of structures, allowing inferences to be made within and between populations by transferring knowledge across them. Within the populations of interest, the similarity of structures, via their corresponding data, should be assessed to successfully implement PBSHM. This paper focusses on using distance metrics to assess similarity at the very start of the analysis chain, to discover information about a population for which there is little prior knowledge and before any analysis has taken place on individual structures. By doing so, it is possible to quickly and automatically identify abnormalities within the population, group similarly behaving structures together, and inform further decisions. The suitability of several candidate metrics that are not widely employed in SHM are tested using a number of commonly occurring feature behaviours, such as varying amplitudes and temporary mean shifts. The effect of data normalisation/standardisation on the metrics is also explored to identify interesting behaviours within the data. A case study is then presented where distance metrics are used to discover similarities and dissimilarities within temperature data from turbines in an offshore wind farm.

Expanding Irreducible Element models: beams and composites

Population-based Structural Health Monitoring's (PBSHM) aspiration is to accumulate a deeper knowledge of a structure's health, by harvesting available data across a range of similar structures (or substructures). Whilst each structure is represented via a unified language called an Irreducible Element (IE) model, the comparison and similarity metrics are computed within a graph space. As such, each model is subsequently converted into an Attributed Graph (AG) before being placed into the PBSHM Framework's comparison graph space; the network. This paper looks into the application of using the PBSHM Framework's comparison graph space as a complex network and determining if the theories of communities and community structures from network science, can be applied to the PBSHM Framework's network, to determine similarity and define yet unknown populations.

Transfer learning via intermediate structures

Recent advances have been made in the field of population-based structural health monitoring (PBSHM), which seeks to share valuable information across a population to improve inferences regarding the health states of the members. However, enabling knowledge transfer between structures with highly disparate features (i.e., heterogeneous populations) is an ongoing challenge. The current work proposes a technique for knowledge transfer within a heterogeneous population, via intermediate structures that help bridge the gap in information between the structures of interest. The transformation from the source to the intermediate and target domains is based on geodesic flows, to incrementally map between changes in geometric and statistical properties. The proposed technique is demonstrated using simulated lumped-mass systems.

An Application of Domain Adaptation for Population-Based Structural Health Monitoring

An Application of Domain Adaptation for Population-Based Structural Health Monitoring

Using the inverse finite‐element method to harmonise classical modal analysis with fibre‐optic strain data for robust population‐based structural health monitoring

AbstractVibration‐based approaches to structural health monitoring (SHM) gained increasing significance for assessing the behaviour of existing structures because of their non‐intrusive nature and high sensitivity to damage. However, data availability often limits the application of SHM approaches. The population‐based structural health monitoring (PBSHM) theory addresses this challenge, enhancing diagnostic inferences by sharing knowledge across a population of similar structures. In real‐life scenarios, sharing data from distinct structures requires dealing with results obtained with different experimental setups, multiple sensors, input choices and acquisition systems. Therefore, it is crucial to harmonise various features to achieve accurate and reliable results. The present study presents the results of a classic experimental modal analysis (EMA) using scanning laser Doppler vibrometer (SLDV) measurements and a strain‐based EMA conducted using high‐definition distributed fibre‐optic strain sensors. The experimental case study of a laboratory‐scale steel aircraft subjected to specific operating and damage conditions is introduced, allowing for a comprehensive discussion of the features extracted from the two EMA techniques, which can also be generalised to structures within different domains. This research highlights the advantages and limitations of fibre‐optic‐based EMA compared to classic methods, as fibre‐optic strain sensors offer a cost‐effective alternative to accelerometers or SLDV for dynamic testing. Furthermore, the feasibility of employing the inverse finite‐element method (iFEM) in the dynamic domain is investigated. This method can estimate the whole displacement field of a structure from a limited number of strain values, thus harmonising strain measurements with the SLDV measurements. By analysing the features extracted from different EMA techniques within the PBSHM framework, this study contributes to advancing the understanding and application of the PBSHM approach in diverse experimental scenarios, laying the foundation for further investigation of features and adequate methods for sharing damage‐state knowledge across a population of structures.

On population-based structural health monitoring for bridges: Comparing similarity metrics and dynamic responses between sets of bridges

Bridges are valuable infrastructure assets that are challenging and expensive to maintain. State-of-the-art data-based bridge SHM solutions look to use bridge response data for condition assessment and damage detection. Data-based SHM methods can be limited in their application as they require large datasets to train models effectively, and most bridges lack the available data for the approaches to work. Further, it would be expensive and unrealistic to collect the required datasets to employ data-based methods to entire bridge networks. Recently, a population-based structural health monitoring (PBSHM) approach was proposed that seeks to leverage the data available for SHM problems by pooling together similar structures with their datasets. The PBSHM approach could be valuable in bridge SHM, enhancing the datasets available for ‘populations of bridges. The PBSHM approach for assessing the similarity of bridges has been considered before and was shown to be useful for identifying similar and different bridge types. However, no data were considered in the previous work, and similarity metrics were only qualified using engineering judgement. For the PBSHM approach to be useful in bridge SHM, there is still a need to check that ‘similar’ bridges have similar responses for transfer learning to be feasible. This paper expands upon previous work and provides originality by investigating if bridges identified as similar also exhibit similar responses. The PBSHM derived similarity metrics convey the topological similarity between structures, and mode shapes are identified as being a topologically sensitive bridge response. Therefore, a modal test campaign is carried out for a set of six real bridges, and Operational Modal Analysis is used to identify modal responses from each of the bridge decks. The Modal Assurance Criterion is used to evaluate the similarity between the mode shapes from pairs of bridges and is subsequently compared to the similarity metrics evaluated between those bridges. The similarity metrics were found to be reflective of the similarities identified between the respective bridges’ mode shapes for bridges of the same and different types. The significance of this finding is that it is an important step towards validating the PBSHM comparison approach for identify similar structures where transfer learning might be attempted.

A domain adaptation approach to damage classification with an application to bridge monitoring

Data-driven machine-learning algorithms generally suffer from a lack of labelled health-state data, mainly those referring to damage conditions. To address such an issue, population-based structural health monitoring seeks to enrich the original dataset by transferring knowledge from a population of monitored structures. Within this context, this paper presents a transfer learning approach, based on domain adaptation, to leverage information from completely-labelled bridge structure data to accurately predict new instances of an unknown target domain. Since intrinsic structural differences may cause distribution shifts, domain adaptation attempts to minimise the distance between the domains and to learn a mapping within a shared feature space. Specifically, the methodology involves the long-term acquisition of natural frequencies from several structural scenarios. Such damage-sensitive features are then aligned via domain adaptation so that a machine-learning algorithm can effectively utilise the labelled source domain data and generalise well to the unlabelled target-domain data. The described procedure is applied to two case studies, including the Z24 and the S101 benchmark bridges and their finite element models, respectively. The results demonstrate the successful exchange of health-state labels to identify the damage class within a population of bridges equipped with SHM systems, showing potential to reduce computational efforts and to deal with scarce or poor data sets in application to bridge network monitoring.

On the hierarchical Bayesian modelling of frequency response functions

Structural health monitoring (SHM) strategies seek to evaluate, predict, and maintain structural integrity, to improve the safety and design service life of structures in operation. Many of these strategies involve monitoring changes in structural dynamics, as damage can affect modal properties and present as changes in the characteristics of the resonance peaks of the frequency response function (FRF). While recent advances have improved the safety and reliability of structures, a number of challenges remain, impeding the practical implementation and generalisation of these systems. Like damage, benign variations, such as those caused by changes in temperature or other environmental fluctuations, can affect dynamic properties, making it difficult to distinguish between damage and normal operating conditions. In addition, newly-deployed structures can have insufficient data to describe the normal operating conditions (i.e., data scarcity), which can impair the development of data-based prediction models. Another common challenge is data loss (i.e., data sparsity), which may result from transmission issues, sensor failure, a sample-rate mismatch between sensors, and other causes. Missing data in the time domain will result in decreased resolution in the frequency domain, which can impair dynamic characterisation.For situations that may benefit from information sharing among datasets, e.g., population-based SHM of similar structures, the hierarchical Bayesian approach provides a useful modelling structure. Hierarchical Bayesian models learn statistical distributions at the population (or parent) and the domain levels simultaneously, to bolster statistical strength among the parameters. As a result, variance is reduced among the parameter estimates, particularly when data are limited. In this paper, a combined probabilistic FRF model is developed for a small population of nominally-identical helicopter blades, using a hierarchical Bayesian structure, to support information transfer in the context of sparse data. The modelling approach is also demonstrated in a traditional SHM context, for a single helicopter blade exposed to varying temperatures, to show how the inclusion of physics-based knowledge can improve generalisation beyond the training data, in the context of scarce data. These models address critical challenges in SHM, by accommodating benign variations that present as differences in the underlying dynamics, while also considering (and utilising), the similarities among the domains.

On the application of domain adaptation for knowledge transfer and damage detection across bridge spans: an experimental case study

Machine Learning offers a substantial support for vibration-based Structural Health Monitoring and automatic damage assessment in real-time. However, labelled health-state data are often limited and difficult to collect, especially under various environmental conditions and damage scenarios. To bridge this gap, Population-Based Structural Health Monitoring is employed to enrich the available dataset by acquiring information from a population of similar structures and to afterwards develop Transfer Learning strategies across different structural systems. In this field, a major challenge is that the development and validation of robust techniques require a large dataset from multiple real structures covering a wide range of health-states. To this end, the present paper firstly illustrates an experimental campaign conducted on a laboratory-scale bridge model to create a comprehensive dataset using different bridge configurations, that are subjected to various environmental conditions and simulated damage scenarios. It follows that the described population certainly provides the basis to investigate TL and gain insights into the field of PBSHM. Working on this case-study, a new application of Domain Adaptation is proposed, with the aim to perform damage identification via label sharing across different spans of a continuously monitored bridge. The extracted natural frequencies are projected into a latent feature space and a machine learning algorithm is then adopted to recognise a certain damage in one span, based on the information previously learned on a different bridge span. This approach is useful if considering that multiple portions could be reasonably affected by similar defects. Therefore, the proposed approach overcomes the limitations of conventional ML-based SHM methods, paving the way to facilitate risk assessment in transportation networks and identify similar anomalies by exchanging span-related health-state information.

Condition transfer between prestressed bridges using structural state translation for structural health monitoring

Implementing Structural Health Monitoring (SHM) systems with extensive sensing layouts on all civil structures is obviously expensive and unfeasible. Thus, estimating the state (condition) of dissimilar civil structures based on the information collected from other structures is regarded as a useful and essential way. For this purpose, Structural State Translation (SST) has been recently proposed to predict the response data of civil structures based on the information acquired from a dissimilar structure. This study uses the SST methodology to translate the state of one bridge (Bridge #1) to a new state based on the knowledge acquired from a structurally dissimilar bridge (Bridge #2). Specifically, the Domain-Generalized Cycle-Generative (DGCG) model is trained in the Domain Generalization learning approach on two distinct data domains obtained from Bridge #1; the bridges have two different conditions: State-H and State-D. Then, the model is used to generalize and transfer the knowledge on Bridge #1 to Bridge #2. In doing so, DGCG translates the state of Bridge #2 to the state that the model has learned after being trained. In one scenario, Bridge #2’s State-H is translated to State-D; in another scenario, Bridge #2’s State-D is translated to State-H. The translated bridge states are then compared with the real ones via modal identifiers and mean magnitude-squared coherence (MMSC), showing that the translated states are remarkably similar to the real ones. For instance, the modes of the translated and real bridge states are similar, with the maximum frequency difference of 1.12% and the minimum correlation of 0.923 in Modal Assurance Criterion values, as well as the minimum of 0.947 in Average MMSC values. In conclusion, this study demonstrates that SST is a promising methodology for research with data scarcity and population-based structural health monitoring (PBSHM). In addition, a critical discussion about the methodology adopted in this study is also offered to address some related concerns.

On the application of population-based structural health monitoring in aerospace engineering.

One of the major obstacles to the widespread uptake of data-based Structural Health Monitoring so far, has been the lack of damage-state data for the (mostly high-value) structures of interest. To address this issue, a methodology for sharing data and models between structures has been developed-Population-Based Structural Health Monitoring (PBSHM). PBSHM works on the principle that, if populations of structures are sufficiently similar, or share sections which can be considered similar, then data and models can be shared between them for use in diagnostic inference. The PBSHM methodology therefore relies on two key components: firstly, identifying whether structures are sufficiently similar for successful transfer of diagnostics; this is achieved by the use of an abstract representation of structures. Secondly, machine learning techniques are exploited to effectively transfer information between the structures in a way that improves damage detection and classification across the whole population. Although PBSHM has been conceived to deal with large and general classes of structures, much of the detailed developments presented so far have concerned bridges; the aim of this paper is to provide similarly detailed discussions in the aerospace context. The overview here will examine data transfer between aircraft components, as well as illustrating how one might construct an abstract representation of a full aircraft.

Modelling variability in vibration-based PBSHM via a generalised population form

Structural health monitoring (SHM) has been an active research area for the last three decades, and has accumulated a number of critical advances over that period, as can be seen in the literature. However, SHM is still facing challenges because of the paucity of damage-state data, operational and environmental fluctuations, repeatability issues, and changes in boundary conditions. These issues present as inconsistencies in the captured features and can have a huge impact on the practical implementation, but more critically, on the generalisation of the technology. Population-based SHM has been designed to address some of these concerns by modelling and transferring missing information using data collected from groups of similar structures. In this work, vibration data were collected from four healthy, nominally-identical, full-scale composite helicopter blades. Manufacturing differences (e.g., slight differences in geometry and/or material properties), among the blades presented as variability in their structural dynamics, which can be very problematic for SHM based on machine learning from vibration data. This work aims to address this variability by defining a general model for the frequency response functions of the blades, called a form, using mixtures of Gaussian processes.

Domain-adapted Gaussian mixture models for population-based structural health monitoring

Transfer learning, in the form of domain adaptation, seeks to overcome challenges associated with a lack of available health-state data for a structure, which severely limits the effectiveness of conventional machine learning approaches to structural health monitoring (SHM). These technologies utilise labelled information across a population of structures (and physics-based models), such that inferences are improved, either for the complete population, or for particular target structures — enabling a population-based view of SHM. The aim of these methods is to infer a mapping between each member of the population’s feature space (called a domain) in which a classifier trained on one member of the population will generalise to the remaining structures. This paper introduces the domain-adapted Gaussian mixture model (DA-GMM) for population-based SHM (PBSHM) scenarios. The DA-GMM, infers a linear mapping that transforms target data from one structure onto a Gaussian mixture model that has been inferred from source data (from another structure). The proposed model is solved via an expectation maximisation technique. The method is demonstrated on three case studies: an artificial dataset demonstrating the approach’s effectiveness when the target domain differs by two-dimensional rotations; a population of two numerical shear-building structures; and a heterogeneous population of two bridges, the Z24 and KW51 bridges. In each case study, the method is shown to provide informative results, outperforming other conventional forms of GMM (where no target labelled data are assumed available), and provide mappings that allow the effective exchange of labelled information from source to target datasets.

On Population-based structural health monitoring for bridges

The maintenance and repair of bridges (and other large scale infrastructure projects) is a major area which could benefit from Structural Health Monitoring technology. Inspections on bridges can take a long time and require many people, and are therefore conducted infrequently. This low frequency of inspection leaves the chance that damage and dangerous critical failures can occur during the long timeframes between inspections. It might even be the case that an inspection fails to identify sub-surface damage. Therefore some form of continuous monitoring is desirable, especially if such systems can reliably detect sub-surface damage. However, the application of SHM to bridges is made challenging by the cost and practicability of obtaining damage-state data for bridges. Over the lifetime of a single bridge, it is hoped that a critical failure will never occur, and only a small number of the possible damage states will occur. It is also unpractical to intentionally damage structures to obtain damage-state data. Population-based structural health monitoring seeks to overcome the obstacle of the limited data available for a single structure, by allowing data to be shared between similar structures. Bridges represent an interesting challenge for PBSHM as each bridge is unique. As such, an assessment of how similar bridges are to each other is required. To provide this assessment, one must develop an abstract representation for each bridge, and using this to perform a comparison. This paper describes the use of a general approach for assessing the similarity of structures, applied to several bridge examples which are representative of common types of bridges, to show that it can be applied in this field.

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

Population-based structural health monitoring (PBSHM) offers a new viewpoint for structural health monitoring (SHM), allowing diagnostic information to be shared across populations of structures. By extending the set of available damage observations, a population-based approach can diagnose damage previously unseen on a structure of interest by leveraging damage information from other structures in the population. These technologies therefore provide significant benefits for making SHM practicable in a variety of industrial settings. It is proposed that PBSHM methodologies must be comprised of tools to assess similarity coupled with algorithms that perform knowledge transfer. The similarity tools are important for identifying whether an SHM task should be attempted for a given population by assessing both the structural similarities between members of a population and similarities in their data spaces. An abstract representation of structures in a graphical domain is presented as an objective way of assessing structural similarity, with distance metrics utilised for assessing data-space similarities. Knowledge transfer is performed using a branch of transfer learning called domain adaptation. By determining if members of a population are similar in a structural and data-space sense, the risk of negative transfer can be reduced; whereby domain adaptation reduces classification performance. This paper demonstrates a PBSHM methodology for transferring knowledge within a heterogeneous population (a group of non-identical structures). Specifically, the PBSHM methodology is shown to transfer localisation labels from a Gnat aircraft wing to an unlabelled Piper Tomahawk aircraft wing dataset, resulting in 100% classification accuracy.