Search

Digital technologies can be used to gather accurate information about the behavior of structural components for improving systems design, as well as for enabling advanced Structural Health Monitoring (SHM) strategies. New avenues for achieving automated and continuous structural assessment are opened up via development of virtualization approaches delivering so-called Digital Twins (DTs), i.e., digital mirrored representations of physical systems relying on fusion of simulation models and real-time monitoring data. In this framework, the main motivation of the work presented in this paper stems from the existing challenges in the implementation and deployment of a real-time predictive framework for virtualization of dynamic systems. Kalman-based filters are usually employed in this context to address the task of joint input-state prediction in structural dynamics. A Gaussian Process Latent Force Model (GPLFM) approach is exploited in this work to construct flexible data-driven a priori models for the unknown inputs, which are then coupled with a combined deterministic-stochastic state-space model of the structural component under study for Kalman-based input-state estimation. The use of Gaussian Process (GP) regression for this task overcomes the limitations of the conventionally adopted random-walk model, thus limiting the necessity of offline user-dependent calibration of this type of data assimilation methods. This paper proposes the use of alternative covariance functions for GP regression in structural dynamics. A theoretical analysis of the GPLFMs linked to the investigated covariance functions is offered. The outcome of this study provides insights into the applicability of each covariance type for GP-based input-state estimation. The proposed framework is validated via an illustrative simulated example, namely a 3 Degrees of Freedom (DOF) system subjected to an array of different loading scenarios. Additionally, the performance of the method is experimentally assessed on the task of joint input-state estimation during testing of a 3D-printed scaled Wind Turbine (WT) blade. [ABSTRACT FROM AUTHOR]

• A novel deep neural network model is proposed, which uses two branches to reconstruct the real and imaginary parts of the signal, solving the key problem that deep neural network models are difficult to extract effective features from compressed signals in the time domain. • A large-scale frequency transformation matrix is decomposed into the multiplication of two small-scale matrices, greatly reducing the number of trainable parameters. • A Fourier Unit is introduced and its feature extraction ability is improved, further enhancing the feature extraction ability of the whole model without significantly increasing the model parameters. • The proposed method is capable of reconstructing vibration responses without relying on sparsity assumptions. Compressive sensing (CS) has become a promising technique for addressing the transmission and storage challenges associated with big data in structural health monitoring. However, classic CS methods require signals to have good sparsity, limiting their performance in low-sparse situations. In order to address this challenge, this paper proposes a deep learning-based CS algorithm for high accuracy reconstruction of structural vibration responses without the assumption of sparsity. The key innovations of the proposed method can be mainly divided into three points. Firstly, a novel deep neural network model is proposed, which uses two branches to reconstruct the real and imaginary parts of the signal in the frequency domain, solving the key problem that deep neural network models are difficult to extract effective features from compressed signals in the time domain. Secondly, a large-scale frequency transformation matrix is decomposed into the multiplication of two small-scale matrices, greatly reducing the number of trainable parameters, and solving the problem that the original frequency transformation block cannot be directly used to extract global frequency-domain features of structural vibration responses due to the large number of parameters. Thirdly, a Fourier Unit is introduced and its feature extraction ability is improved, further enhancing the feature extraction ability of the whole model without significantly increasing the model parameters. The model does not use the adversarial training strategy, and the training process is simple and stable. It also does not require a large amount of adjustment of hyper-parameters. The number of trainable parameters of the model is only about 4 million, and the model training can be carried out with low-cost hardware. The effectiveness of the proposed method is verified by the low sparsity vibration response of a laboratory grandstand simulator. The maximum sparsity level of the vibration response of the simulator with 30 nodes is 0.381, which is larger than the sparsity level of the vibration response of common large-scale civil structures, that is, the sparsity is smaller and the reconstruction difficulty is greater. Experimental results show that the reconstruction error of the two classic CS algorithms is more than 3 times higher than that of the proposed method across all compression ratios. The identification results of the modal parameters of the original response and the reconstructed response show excellent consistency in their frequencies and vibration modes. The maximum relative error of the first 12 frequencies is only 0.038%, and the modal assurance criterion of the first 12 mode shapes is at least 0.9962. Overall, in large-scale low-sparse situations, the proposed method has high accuracy and high computational efficiency in reconstructing structural vibration responses, which can greatly alleviate the transmission and storage problems of monitoring big data with very low hardware requirements. [ABSTRACT FROM AUTHOR]

Structural identification is an essential process in structural health monitoring, condition assessment and structural safety evaluation. This inverse problem becomes more challenging when information on the dynamic loads is missing or not fully known. Hence, it is important to establish methods to identify structural parameters and dynamic loads simultaneously from the measured structural responses which are easy to obtain compared to dynamic loads. This paper proposes a novel method to identify simultaneously structural parameters and dynamic loads from structural responses measured on a limited set of degrees-of-freedom. Firstly, an objective function is defined as the difference between the measured structural responses and the theoretically computed responses, and then the derivative of the residual function with respect to structural parameters is calculated numerically using the forward-difference method. The derivative of the residual function with respect to the external dynamic loads is computed using the state-space formulation and the system matrix composed of Markov parameters to facilitate the derivative-based identification. Secondly, the nonlinear optimization problem is solved using the Levenberg–Marquardt algorithm. Several numerical examples are analysed to demonstrate the effectiveness and robustness of the method. Finally, the effect of initial estimates of the parameters and dynamic loads and the effect of measurement noise, as well as the effects of number of measurements are investigated. The proposed method is also shown to achieve a satisfactory solution even when the initial estimates of parameters and dynamic loads are far from their true values. • The structural parameters and the dynamic loads are identified simultaneously. • A state-space formulation is used to facilitate the identification. • The Levenberg–Marquardt algorithm is used to solve the optimization problem. • The effects of noise, number of measurements and excitations are analysed. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

In recent years, there has been a growing interest in the development of ground-based Synthetic Aperture Radars (GBSAR) for the purpose of monitoring structural displacements. GBSAR offers high-resolution monitoring over a wide area and can capture data every few minutes. However, compact high-frequency multiple input multiple output (MIMO) radars have emerged as an alternative for monitoring sub-second displacements, such as structural vibrations. MIMO radar has sub-second acquisition interval. However, it has limited cross-range resolution compared to GBSAR, and interference between antennas and presence of multiple scatterers in the scene can cause strong sidelobes in the processed data. On the other hand, GBSAR utilizes a long synthetic aperture to achieve high cross-range resolution. However, due to its longer data acquisition time compared to MIMO radar, conventional methods are insufficient for detecting scatterers' sub-second displacements that occur during the data acquisition process. This study proposes a method to effectively monitor sub-second or sub-minute displacements using GBSAR signals. The proposed method enhances the conventional radar interferometric processes by employing spectral estimation, allowing for multi-dimensional detection of targets' azimuth angle, linear displacement, and vibrational characteristics. Consequently, this method improves both the processing of MIMO radar data and enables high-resolution fast displacement monitoring from GBSAR signals. The paper presents the theoretical details and mathematical formulations of the proposed method for both MIMO radar and GBSAR imaging modes. To evaluate the effectiveness of the proposed method, numerical simulations and real experiments are conducted. The experimental results validate the capability of the proposed method in both GBSAR and MIMO configuration modes for high-resolution monitoring of fast linear displacements and vibrations. The results exhibit promising signal-to-noise ratio (SNR) and peak-to-sidelobe ratio (PSLR) values. [ABSTRACT FROM AUTHOR]

The use of Machine Learning (ML) has rapidly spread across several fields of applied sciences, having encountered many applications in Structural Dynamics and Vibroacoustic (SD&V). An advantage of ML algorithms compared to traditional techniques is that physical phenomena can be modeled using only sampled data from either measurements or simulations. This is particularly important in SD&V when the model of the studied phenomenon is either unknown or computationally expensive to simulate. This paper presents a survey on the application of ML algorithms in three classical problems of SD&V: structural health monitoring, active control of noise and vibration, and vibroacoustic product design. In structural health monitoring, ML is employed to extract damage-sensitive features from sampled data and to detect, localize, assess, and forecast failures in the structure. In active control of noise and vibration, ML techniques are used in the identification of state-space models of the controlled system, dimensionality reduction of existing models, and design of controllers. In vibroacoustic product design, ML algorithms can create surrogates that are faster to evaluate than physics-based models. The methodologies considered in this work are analyzed in terms of their strength and limitations for each of the three considered SD&V problems. Moreover, the paper considers the role of digital twins and physics-guided ML to overcome current challenges and lay the foundations for future research in the field. • Review of machine learning applied in vibroacoustic and structural dynamics. • Basic principles of the machine learning methods commonly used are presented. • Control, structural health monitoring and design are the most relevant applications. • The current implementation scenario and challenges are reviewed for each application. • This article sheds light on trends and research opportunities merging both fields. [ABSTRACT FROM AUTHOR]

Person identification is important in providing personalized services in smart buildings. Many existing studies focus on closed-world person identification, which only identifies a fixed group of people who have training data; however, they assume everyone has pre-collected data, which is not practical in real-world scenarios when newcomers are present. To overcome this drawback, open-world person identification recognizes both newcomers and registered people, which opens up new opportunities for smart building applications that involve newcomers, such as smart visitor management, customized retail, personalized health monitoring, and public emergency assistance. To achieve this, structural vibration sensing has various advantages when compared with the existing sensing modalities (e.g., cameras, wearables, and pressure sensors) because it only needs sparsely deployed sensors mounted on the floor, does not require people to carry devices, and is perceived as more privacy-friendly. However, one fundamental challenge in analyzing footstep-induced structural vibration data is its high variability due to the structural heterogeneity and the footstep variations. Therefore, it is difficult to distinguish different people given this high variability within each person, and it is more challenging to recognize a new person as that data is unobserved before. In this paper, we characterize the variability in footstep-induced structural vibration to develop an open-world person identification framework. Specifically, we address three variability challenges in developing our method. First, the high variability within each person comes from multiple sources that are entangled in the vibration signals, and thus is difficult to be decomposed and reduced. Secondly, the distribution of features extracted from the vibration signals is irregularly shaped, and therefore is difficult to model. Moreover, the identity of the next person is correlated with the previous observations, which makes the identification process more complicated. To overcome these challenges, we first characterize multiple variability sources and design a transformation function that results in signal features that are less variable within one person and more separable between different people. We then develop a modified Chinese Restaurant Process (mCRP) for nonparametric Bayesian modeling to capture the irregularly shaped feature patterns both from local and global perspectives. Finally, we design an adaptive hyperparameter α that represents the prior probability of newcomers at each observation, which keeps updating depending on the time, location, and previous predictions. We evaluate our approach through walking experiments with 20 people across 2 different structures. With only 1 pre-recorded person at each structure, our method achieves up to 92.3% average accuracy with randomly appearing newcomers. [ABSTRACT FROM AUTHOR]

Highlights from the article: It is with pleasure that the Guest Editors introduce this Special Issue for the Journal of Mechanical Systems and Signal Processing (MSSP). The twenty five papers of this Special Issue cover a wide range of interesting issues and have been grouped according to the paper main topics. As in previous editions of ICEDyn, the presented papers are of a very high scientific standard where the authors produce improved versions of their papers especially to this Special Issue.

• A novel algorithm for joint input-state-parameter estimation is proposed. • A delay is introduced in the estimation to reduce estimation errors. • Analytical expressions for the required sensitivities of the system matrices are derived. • The methodology is validated using data obtained from a laboratory experiment. This paper presents a novel algorithm for joint input-state-parameter estimation in structural dynamics. The algorithm is derived from an existing smoothing algorithm. In each time step, the system model adopted in the joint input-state-parameter estimation is linearized around the current state, yielding an algorithm similar to the extended Kalman filter. It is shown that adopting a time delay in the estimation can significantly reduce the estimation error, especially in case data originates from sensors that are not collocated with the estimated inputs. Analytical expressions for the sensitivities of the system matrices with respect to unknown parameters are derived for the case of a linear underlying state-space model. These sensitivities are derived for models expressed in physical coordinates, models expressed in modal coordinates, and modally reduced-order models with a quasi-static correction to account for the contribution of the out-of-band modes. The proposed methodology is verified using numerical simulations and validated using data obtained from a laboratory experiment on a steel beam with I-shaped cross section. [ABSTRACT FROM AUTHOR]

• Compressed Sensing is applied to full-field optical vibration data. • A combined Compressed Sensing/Shape Descriptor method is presented. • Huge data reduction is achieved for purposes of transmission and storage. • Operational modal analysis is demonstrated using recovered full-field data. • Application of the methodology is demonstrated using industrial DIC data. Optically-acquired data, typically from digital image correlation, is increasingly being used in the area of structural dynamics, particularly modal testing and damage identification. One of the problems with such data is its extremely large size. Single images regularly extend to tens or even hundreds of thousands of data points and many thousands of images may be required for a vibration test. Such data must be stored and transmitted efficiently for later remote reconstruction and analysis, typically operational modal analysis. It is this requirement that is addressed in the research presented in this paper. This research builds upon previous work whereby digitised optical data was projected onto an orthogonal basis with coefficients (shape descriptors) of either greater or lesser significance; those deemed to be insignificant, according to a chosen threshold being removed. Data reduction by a combination of shape-descriptor decomposition and compressed-sensing is applied to an industrial printed circuit board and reconstructed for operational modal analysis by ℓ 1 optimisation. [ABSTRACT FROM AUTHOR]

This paper presents a recursive Bayesian inference framework for digital twinning of an offshore wind turbine (OWT) in the Block Island Wind Farm (BIWF), the first commercial offshore wind farm in the U.S. with jacket substructure. The framework employs a window-based extended Kalman filter (EKF) to jointly estimate the dynamic input loads and unknown model parameters of the finite element (FE) model of the turbine using sparse output-only measurements. The estimation of input loads on the OWT is critical for fatigue life estimation, foundation monitoring and control strategy. In the proposed joint estimation algorithm, unknown input time histories are divided into overlapping windows and the current window of inputs is concatenated with uncertain model parameters to form an augmented state vector. To account for the uncertainty about wind direction and misalignment between wind direction and the turbine yaw angle, two independent horizontal wind load time histories in X and Y directions are estimated at the nacelle level. A numerical validation study is first performed in which wind loads of the OWT are simulated using the multi-physics platform FAST, and structural vibration responses are generated using the FE model in OpenSees. The window-based EKF is implemented using the simulated measurements, and the estimations of structural stiffness as the updating model parameter and input loads are observed to have high accuracy. The proposed algorithm is then validated using in-situ measurements on one of the OWT in BIWF. To investigate the effectiveness and robustness of the method under different environmental and operational conditions, four cases are considered in which the turbine is operating at the rated rotor speed (region 3 of power generation), below the rated rotor speed (region 2), idling rotor state (region 1 or no power), and region 4 beyond cut-out wind speed. Accurate predictions of acceleration and strain data are obtained using the proposed algorithm, and the estimations of model parameter are consistent and close to their nominal value under different conditions. These results from numerical and experimental studies validate the proposed window-based EKF for joint parameter-input estimation and digital twinning for long-term structural health monitoring of the OWT. [ABSTRACT FROM AUTHOR]

This paper presents a novel Optimal Sensor Placement (OSP) strategy that is dedicated to model updating problems based on the modified Constitutive Relation Error (mCRE) functional in low-frequency dynamics. The mCRE is a credible alternative to model updating functionals that stands out by searching structural parameters alongside mechanical fields as the best trade-off between all available information from measured data, without any further a priori assumption. Considering damage detection problems, due to possible discrepancies in terms of parameters sensitivity with respect to mCRE, sensor locations provided by standard OSP algorithms may be irrelevant. The proposed approach uses the concept of Information Entropy by formulating a modified Fisher information matrix, in which the sensitivity of the mCRE mechanical fields with respect to the updated parameters is involved. The approach is legitimated by the strong connection between mCRE and Bayesian inference. A proof-of-concept involving an earthquake engineering inspired academic case study, where accelerometers are positioned on a two-story frame structure subjected to random ground motion, permits to illustrate the soundness and efficiency of the proposed methodology compared to other classical OSP techniques. The influence of critical mCRE parameters is shown, as well as the benefits of taking multiple scenarios into account so as to get an OSP that is relevant for a wider range of possible damage occurrences. • A new sensor placement technique is proposed for structural monitoring in dynamics. • It aims to improve model updating based on the modified Constitutive Relation Error. • The sensor placement criterion is obtained by coupling mCRE and information entropy. • The proposed approach is compared to classical techniques on a 3D example. • The relevance of the approach for mCRE-based model updating is illustrated. [ABSTRACT FROM AUTHOR]

This paper suggests a Structural Health Monitoring (SHM) method for damage detection and localization in pipeline. The baseline signals, used in SHM, could change due to the variation of environmental and operational conditions (EOCs). Hence, the damage detection method could give rise to false alarm. In this study, this issue is addressed by estimating the current signal using only few reference signals with similar or very close EOCs. Such an operation can be performed by calculating a sparse estimation of the current signal. The estimation error is used as an indication of the presence of damage. Actually, a signal from the damaged pipe will be characterized by a high estimation error compared to that of a signal from the undamaged pipe. The damage location is obtained by calculating the estimation error on a sliding window over the signal from the damaged pipe. This method was tested on signals collected on a 6 m pipeline segment placed in a workshop under natural temperature variations. Results have shown that the created damage was successfully detected and localized. [ABSTRACT FROM AUTHOR]

The subject of structural health monitoring is drawing more and more attention over the last years. Many vibration-based techniques aiming at detecting small structural changes or even damage have been developed or enhanced through successive researches. Lately, several studies have focused on the use of raw dynamic data to assess information about structural condition. Despite this trend and much skepticism, many methods still rely on the use of modal parameters as fundamental data for damage detection. Therefore, it is of utmost importance that modal identification procedures are performed with a sufficient level of precision and automation. To fulfill these requirements, this paper presents a novel automated time-domain methodology to identify modal parameters based on a two-step clustering analysis. The first step consists in clustering modes estimates from parametric models of different orders, usually presented in stabilization diagrams. In an automated manner, the first clustering analysis indicates which estimates correspond to physical modes. To circumvent the detection of spurious modes or the loss of physical ones, a second clustering step is then performed. The second step consists in the data mining of information gathered from the first step. To attest the robustness and efficiency of the proposed methodology, numerically generated signals as well as experimental data obtained from a simply supported beam tested in laboratory and from a railway bridge are utilized. The results appeared to be more robust and accurate comparing to those obtained from methods based on one-step clustering analysis. [ABSTRACT FROM AUTHOR]

The establishment of a Digital Twin of an operating engineered system can increase the potency of Structural Health Monitoring (SHM) tools, which are then bestowed with enhanced predictive capabilities. This is particularly relevant for wind energy infrastructures, where the definition of remaining useful life is a main driver for assessing the efficacy of these systems. In order to ensure a proper representation of the physical structure, the monitored response of the Digital Twin should match the one experienced by the actual system throughout the complete spectrum of its operational conditions. In most typical SHM configurations, it is only possible to rely on output-only measurements, acquired from finite positions within a structure, which naturally raises the challenge of recovering the full-field operational response, including unmeasured locations. This problem, also known as Virtual Sensing (VS), has been treated using different schemes, including Bayesian filtering and Modal Expansion (ME). In this paper, the Augmented Kalman Filter (AKF) is exploited to this end; a tool which allows for simultaneous full-field response and unmeasured input prediction. The common issue of Bayesian filtering relies on calibration of the filters defining parameters, namely the assumed measurement and process noise covariance levels. While the first is directly related to the accuracy of the employed physical sensors, the latter often acts as a tuning parameter for improving the reliability of the prediction. The process noise covariance adjustment is often performed in an offline fashion, either by making use of regularization methods, e.g., the L-curve method, or via trial and error. In this work, we propose a methodology for automated process noise covariance adaptation, relying on response estimates recovered by means of an improved ME approach. The method is validated on experimental data from a large scale research Wind Turbine (WT) blade made of glass fiber reinforced plastics. [ABSTRACT FROM AUTHOR]

• An enhanced method for reducing reconstruction error of the CS is proposed. • Prior information is provided by the time series feature of vibration signal. • The optimization problem is formulated to decrease the residuum of the AR model. • The compressed measurement is transformed as the general solution of nonhomogeneous linear equation. Compressed sensing (CS) utilizes the signal's sparsity to reconstruct signals from far less linear measurements. However, ambient vibration response in structural dynamics typically lacks sparsity on a regular transform basis. Hence, when the vibration signals are reconstructed through the CS, significant errors are unavoidably induced, especially at high compression ratios, limiting the CS applicability in structural health monitoring and damage detection. To address these issues, this paper proposes an enhanced error reduction method, exploited as a post-processing scheme for signal reconstruction. The suggested method constructs an autoregression model, whose residuum increases to correspond to the reconstruction error based on empirical observations. Through minimizing the residuum under the constraint of compressed measurements, the reconstruction error is then minimized, which leads to an optimized result of the reconstructed signals. The suggested method is validated using ambient vibration data collected from a laboratory-scale shear frame model and a full-scale cable-stayed bridge. [ABSTRACT FROM AUTHOR]

Abstract: The failure of critical components in industrial systems may have negative consequences on the availability, the productivity, the security and the environment. To avoid such situations, the health condition of the physical system, and particularly of its critical components, can be constantly assessed by using the monitoring data to perform on-line system diagnostics and prognostics. The present paper is a contribution on the assessment of the health condition of a computer numerical control (CNC) tool machine and the estimation of its remaining useful life (RUL). The proposed method relies on two main phases: an off-line phase and an on-line phase. During the first phase, the raw data provided by the sensors are processed to extract reliable features. These latter are used as inputs of learning algorithms in order to generate the models that represent the wear''s behavior of the cutting tool. Then, in the second phase, which is an assessment one, the constructed models are exploited to identify the tool''s current health state, predict its RUL and the associated confidence bounds. The proposed method is applied on a benchmark of condition monitoring data gathered during several cuts of a CNC tool. Simulation results are obtained and discussed at the end of the paper. [Copyright &y& Elsevier]

Abstract: Frequency analysis is often used in structural dynamics, and the precautions and requirements necessary to obtain accurate results in the frequency domain are well known. However there are applications where resources limitations mean that full accuracy has to be compromised; one example is structural health monitoring using wireless sensor systems with severe constraints on power consumption and processing power. This paper investigates two approximations that may be introduced into the Discrete Fourier Transform computation, namely the quantization of the kernel function (the mono-bit FFT) and the quantization of the time signal, and considers the errors introduced into the frequency spectrum. The paper demonstrates that surprising good results may be obtained and these approximate methods should be considered during the design of systems constrained by resource limitations. [Copyright &y& Elsevier]

Abstract: The use of damage-sensitive features to evaluate structural condition or health is a very critical aspect of structural health monitoring. The purpose of this paper is to investigate the potential of two different damage-sensitive features for detecting damage. Different damage scenarios are simulated on a large-scale laboratory structure and a three-span highway bridge for demonstration. The features presented in this paper are the modal flexibility-based deflection and curvature both of which are obtained directly from dynamic properties. In the literature, flexibility associated with mode shapes and mode shapes curvatures have been mostly explored. In this study, multi-input–multi-output dynamic data are used to obtain modal flexibility, which is a close approximation to the actual flexibility. A main novelty is that the curvature is calculated from the deflected shapes using the modal flexibility as opposed to using modal vectors. In this paper, the theory of the methodology is explained and then experimental studies and results are presented. For the experimental studies, the laboratory specimen and the three-span bridge were gradually damaged. It is shown that both deflection and curvature are conceptual and physically meaningful features for damage detection and localization. The issues and the requirements for these features to perform successfully are also presented. [Copyright &y& Elsevier]

Abstract: This paper presents the application of a non-linear feature identification technique for structural damage detection. This method is coupled with the impedance-based structural health monitoring (SHM) method, which utilises electromechanical coupling properties of piezoelectric materials. The non-linear feature examined in this study is in the form of autoregressive coefficients in the frequency domain autoregressive model with exogenous (ARX) inputs, which explicitly considers non-linear system input/output relationships. The applicability of this non-linear feature for damage identification is investigated in various frequency ranges using impedance signals measured from a laboratory-test structure. The performance of the non-linear feature is also compared with those of linear features typically used in impedance methods. This paper reinforces the utility of non-linear features in SHM and suggests that their sensitivity in different frequency ranges may be leveraged for certain applications. [Copyright &y& Elsevier]

Randomly missing data of structural vibration responses time history often occurs in structural dynamics and health monitoring. For example, structural vibration responses are often corrupted by outliers or erroneous measurements due to sensor malfunction; in wireless sensing platforms, data loss during wireless communication is a common issue. Besides, to alleviate the wireless data sampling or communication burden, certain accounts of data are often discarded during sampling or before transmission. In these and other applications, recovery of the randomly missing structural vibration responses from the available, incomplete data, is essential for system identification and structural health monitoring; it is an ill-posed inverse problem, however. This paper explicitly harnesses the data structure itself—of the structural vibration responses—to address this (inverse) problem. What is relevant is an empirical, but often practically true, observation, that is, typically there are only few modes active in the structural vibration responses; hence a sparse representation (in frequency domain) of the single-channel data vector, or, a low-rank structure (by singular value decomposition) of the multi-channel data matrix. Exploiting such prior knowledge of data structure (intra-channel sparse or inter-channel low-rank), the new theories of ℓ 1 -minimization sparse recovery and nuclear-norm-minimization low-rank matrix completion enable recovery of the randomly missing or corrupted structural vibration response data. The performance of these two alternatives, in terms of recovery accuracy and computational time under different data missing rates, is investigated on a few structural vibration response data sets—the seismic responses of the super high-rise Canton Tower and the structural health monitoring accelerations of a real large-scale cable-stayed bridge. Encouraging results are obtained and the applicability and limitation of the presented methods are discussed. [ABSTRACT FROM AUTHOR]

• For the first time a new population-based methodology is proposed that alleviates problems with sparsity of data in SHM. • The applicability of domain adaptation assumptions when considering each population type and SHM problem are outlined. • A novel MMD-based algorithm is proposed for an (L + 1)-problem. • A new approach for using physics-based models and domain adaption is proposed for labelling data from structures in PBSHM. • This is the third paper (of three parts) covering the foundations of population-based SHM. This is the third and final paper in a series laying foundations for a theory/methodology of Population-Based Structural Health Monitoring (PBSHM). PBSHM involves utilising knowledge from one set of structures in a population and applying it to a different set, such that predictions about the health states of each member in the population can be performed and improved. Central ideas behind PBSHM are those of knowledge transfer and mapping. In the context of PBSHM, knowledge transfer involves using information from a source domain structure, where labels are known for given feature sets, and mapping these onto the unlabelled feature space of a different, target domain structure. This mapping means a classifier trained on the transformed source domain data will generalise to the unlabelled target domain data; i.e. a classifier built on one structure will generalise to another, making Structural Heath Monitoring (SHM) cost-effective and applicable to a wide range of challenging industrial scenarios. This process of mapping features and labels across source and target domains is defined here via domain adaptation , a subcategory of transfer learning. A mathematical underpinning for when domain adaptation is possible in a structural dynamics context is provided, with reference to topology within a graphical representation of structures. Subsequently, a novel procedure for performing domain adaptation on topologically different structures is outlined. [ABSTRACT FROM AUTHOR]

Damage diagnosis based on global structural vibrations critically depends on the sensor layout, in particular when a small number of sensors is used for large structures under unknown excitation. This paper proposes a sensor placement strategy that yields an optimized sensor layout with maximum damage detectability in selected structural components. The optimization criterion is based on the Fisher information, which quantifies the information that the damage-sensitive feature carries on the design parameters of structural components, such as material constants or cross-sectional values. It is evaluated using a finite element model, and considers the statistical uncertainties of the damage-sensitive feature. The methodology is shown for the stochastic subspace-based damage detection method, but can be applied to any damage-sensitive feature whose distribution can be approximated as Gaussian. It is suitable to find the optimal layout for a fixed number of sensors and to choose an appropriate number of sensors. Since the Fisher information is defined component-wise, the sensor layout can be tuned to become more sensitive to damage in local structural components, such as damage hotspots, non-inspectable components, or components that are critical for the safety and serviceability of the structure. For proof of concept, the sensor layout on a laboratory beam is optimized based on numerical simulations, and it is showcased that the optimal sensor layout leads to the highest damage detectability for experimental data. • Gaussian vibration-based features are employed for statistical damage detection. • The sensor layout is optimized to maximize the feature's sensitivity and minimize its uncertainty. • The optimization criterion allows one to tune the sensor layout for high detectability in critical components. • A custom-made genetic algorithm efficiently finds a close-to-optimal solution for large structures. • The optimization is based on a numerical model and validated in a lab experiment. [ABSTRACT FROM AUTHOR]

Abstract: Bridge cables, characterized by distributed large span, serving in harsh environment and vulnerability to random damage, are the key load-sustaining components of cable-based bridges. To ensure the safety of the bridge structure, it is critical to monitor the loading conditions of these cables under lengthwise random damages. Aiming at obtaining accurate monitoring at the critical points as well as the general information of the cable force distributed along the entire cable, this paper presents a study on cable force monitoring by combining optical fiber Bragg grating (FBG) sensors and Brillouin optical time domain analysis/reflectory (BOTDA/R) sensing technique in one single optical fiber. A smart FRP-OF-FBG rebar based cable was fabricated by protruding a FRP packaged OF-FBG sensor into the bridge cable. And its sensing characteristics, stability under high stress state temperature self-compensation as well as BOTDA/R distributed data improvement by local FBG sensors have been investigated. The results show that FRP-OF-FBG rebar in the smart cable can deform consistantly along with the steel wire and the cable force obtained from the optical fiber sensors agree well with theoretical value with relative error less than ±5%. Besides, the temperature self-compensation method provides a significant cost-effective technique for the FRP-OF-FBG based cables'' in situ cable force measurement. And furthermore, potential damages of the bridge cable, e.g. wire breaking and corrosion, can be characterized and symbolized by the discontinuity and fluctuation of the distributed BOTDA data thereafter accuracy improved by local FBG sensors. [Copyright &y& Elsevier]

Abstract: A sound understanding of a structure''s normal condition, including its response to normal environmental and operational variations is desirable for structural health monitoring and necessary for performance monitoring of civil structures. The current paper outlines the extensive monitoring campaign of the Tamar Suspension Bridge as well as analysis carried out in an attempt to understand the bridge''s normal condition. Specifically the effects of temperature, traffic loading and wind speed on the structure''s dynamic response are investigated. Finally, initial steps towards the development of a structural health monitoring system for the Tamar Bridge are addressed. [Copyright &y& Elsevier]

Abstract: Continuous monitoring of structural vibrations is becoming increasingly common as sensors and data acquisition systems become more affordable, and as system and damage identification methods develop. In vibration-based structural health monitoring, the dynamic modal parameters of a structure are usually used as damage-sensitive features. The modal parameters are often sensitive to changing environmental conditions such as temperature, humidity, or excitation amplitude. Environmental conditions can have as large an effect on the modal parameters as significant structural damage, so these effects should be accounted for before applying damage identification methods. This paper presents results from a continuous monitoring system installed on the Dowling Hall Footbridge on the campus of Tufts University. Significant variability in the identified natural frequencies is observed; these changes in natural frequency are strongly correlated with temperature. Several nonlinear models are proposed to represent the relationship between the identified natural frequencies and measured temperatures. The final model is then validated using independent sets of measured data. Finally, confidence intervals are estimated for the identified natural frequencies as a function of temperature. The ratio of observed outliers to the expected rate of outliers based on the confidence level can be used as a damage detection index. [Copyright &y& Elsevier]

Abstract: Modal-based damage detection algorithms are well-known techniques for structural health assessment, but they are not commonly used due to the lack of automated modal identification and tracking procedures. Development of such procedures is not a trivial task since traditional modal identification requires extensive interaction from an expert user. Nevertheless, computational efforts have to be carefully considered. If fast on-line data processing is crucial for quickly varying in time systems (such as a rocket burning fuel), a number of vibration-based condition monitoring applications are performed at very different time scales, resulting in satisfactory time steps for on-line data analysis. Moreover, promising results in the field of automated modal identification have been recently achieved. In the present paper, a literature review on this topic is presented and recent developments concerning fully automated output-only modal identification procedures are described. Some case studies are also reported in order to validate the approach. They are characterized by different levels of complexity, in terms of mode coupling, dynamic interaction effects and level of vibration. Advantages and drawbacks of the proposed approach will be pointed out with reference to available experimental results. The final objective is the implementation of a fully automated system for vibration-based structural health monitoring of civil engineering structures and identification of adequate requirements about sensor number and layout, record duration and hardware characteristics able to ensure a reliable low-cost health assessment of constructions. Results of application of the proposed methodology to modal parameter estimation in operational conditions and during ground motions induced by the recent L’Aquila earthquake will be finally presented and discussed. [Copyright &y& Elsevier]

• Proposing a new identifiable hysteretic model with explicit physical parameters determined by hysteresis loop. • Extracting dynamic features for training SVM model to identify unknown parameters. • Combining SHM and SVM to automatically create a nonlinear model for response prediction. • Validating the method against a full scale 3-story real-world building structure with realistic nonlinear seismic response. • Extending SHM from a retrospective monitoring tool to a prospective tool in earthquake response and recovery. Structural health monitoring (SHM) is backwards analysis of past to current state of damage, but cannot create a structure-specific nonlinear model for forward analysis of future response and damage. This paper aims to develop an automated modelling approach to translate proven hysteresis loop analysis (HLA) SHM results into nonlinear foundation models for response forecasting in subsequent events, particularly for steel structures with post-yielding behaviors. Support vector machine (SVM) is employed to identify the proposed nonlinear baseline model. Stiffness features are extracted from HLA to train the SVM model incorporating the constraints of SHM identification. A proof-of-concept case study validates the ability of the proposed method to accurately identify 12 model parameters with average error of 2.8% for a nonlinear numerical structure in the presence of 10% RMS measurement noise. Experimental validation from a full-scale 3-storey real building shows the predicted nonlinear responses match the measured response well with cross correlation coefficients R coeff = 0.94, 0.92 and 0.89 for the first, second and third floor, respectively. In addition, the predicted stiffness changes also match the SHM results very well with errors less than 2.1%. Finally, and most importantly, the identified model is able to predict the response of 2 further events with average of correlation coefficient R coeff = 0.91 and average error of 1.9% for stiffness changes across all cases. The overall results validate the ability of the created predictive model to accurately capture the essential dynamics and structural degradation, as well as predicting future possible response and risk. [ABSTRACT FROM AUTHOR]

• Event cameras provide full-field structural monitoring and vibration analysis. • Learned structural response function is highly physically interpretable. • The proposed framework enables structural boundary condition learning. This paper exploits a new direction of full-field structural monitoring and vibration analysis, using an emerging type of neuro-inspired vision sensors – namely event cameras. Compared to traditional frame-based cameras, event cameras offer salient benefits of resilience to motion blur, high dynamic range, and microsecond latency. Event cameras are herein exploited for structural monitoring, in order to extract dense measurements of structural response in terms of both spatial and temporal resolution. The output of an event camera is a stream of so called "events", which is different to traditional snapshots. Due to this fundamentally different working principle, basic computer vision algorithms, such as optical flow or feature tracking, should be re-designed for processing event-based measurements. In this work, we present a novel framework termed physics-informed sparse identification, for full-field structural vibration tracking and analysis. The framework leverages sparse identification guided by assimilation of the underlying structural dynamics in the assembly of a library matrix, which is used to characterize the system's dynamics. The stream of event data generated from event cameras is sparsely represented by means of well-chosen basis functions, allowing for a physical interpretation of the system's response. The proposed framework is extended to boundary condition learning/classification by fusion of characteristic basis functions, representing different classes of support conditions, into the library matrix. The results obtained by means of an illustrative numerical example, as well as experimental tests on vibrating beams recorded by an event camera demonstrate an accurate tracking of structural vibration and the developed strains, in the form of full-field measurements rather than point-wise tracking. What is more, the proposed sparse learning process enables identification of the boundary conditions of monitored structural elements, which comes with key benefits for structural monitoring. [ABSTRACT FROM AUTHOR]