Bayesian‐optimized unsupervised learning approach for structural damage detection

The frequency-shift-based scheme is one of the popular structural damage detectionmethods used in the field. Assuming that the damage will cause structural parameter(e.g. mass or stiffness) changes, one can detect damage by observing natural frequencyvariations before and after damage occurs. One of the major limitations of thefrequency-shift-based damage detection approach is that the natural frequencies could beinsensitive to local and small parameter variations (damages) in the structure. To addressthis problem, the concept of sensitivity-enhancing control has been developed andexplored by researchers. Recently, an eigenstructure (dual eigenvalue–eigenvector)assignment-based approach has been proposed to advance the state of the artof sensitivity-enhancing control design for multiple degrees-of-freedom systems(Jiang et al 2007 Trans. ASME, J. Vib. Acoust. 129 771–83). While promisingfeatures have been shown, the approach has not been experimentally verified. Inaddition, such a scheme requires an accurate analytical model, which is in generaldifficult to obtain for complex structures. The purpose of this research is to advancethe state of the art by exploring an experimental-based sensitivity-enhancingcontrol approach for structural damage detection. That is, we aim at utilizingmathematical models identified from experimental measurement data for controllerdesign and system analysis, and hence eliminate the requirement of an accurateanalytical model. With the identified model, an eigenstructure assignment-basedoptimization scheme is utilized to design the closed-loop control to best enhance thenatural frequency sensitivity to structural damages. An experimental investigationbased on a test bed beam structure is conducted to demonstrate and verify theproposed method. Sensitivity-enhancing controls are designed to increase thefirst two natural frequency sensitivities to mass changes in the beam structure.Experimental results show that by applying the designed sensitivity-enhancing controls tothe beam structure, the natural frequency sensitivities to mass variations canbe significantly enhanced. Results also show that the damage locations can besuccessfully identified and the damage severity can be predicted by using the measuredfrequency changes of the closed-loop systems before and after damage occurrence.

It is still necessary to investigate the detection of structural damage under ambient excitations since the excitations are random and unmeasured while measurement noises are inevitable. In this paper, a method based on the synthesis of cross-correlation functions of partial structural responses and the extended Kalman filter (EKF) approach is proposed for the identification and damage detection of structures under ambient excitations, in which both independent stationary and non-stationary white noise excitations in the product models are discussed. First, the equations of cross-correlation functions of structural responses are established when the ambient excitations are independent stationary white noise processes. Then, the EKF approach is utilized to identify structural parameters and cross-correlation functions using partial measurements of structural acceleration responses. Structural damage is detected based on the degradations of the identified structural element stiffness parameters. Finally, the proposed method is extended to deal with independent non-stationary white noise excitations in the product models. The numerical simulation examples of the ASCE structural health monitoring benchmark building subject to ambient excitation, a moment resisting frame model under white noise excitation, and a cantilever beam model under multiple independent non-stationary excitations are used to validate the feasibility of the proposed method. It is shown that the method is not sensitive to measurement noises. Also, a lab experimental study of the identification of a multi-story shear structure is investigated to further illustrate the applicability of the proposed method.

The high structural deficient rate poses serious risks to the operation of many bridges and buildings. To prevent critical damage and structural collapse, a quick structural health diagnosis tool is needed during normal operation or immediately after extreme events. In structural health monitoring (SHM), many existing methods will have limited usefulness in the quick damage identification process because (1) the damage event needs to be identified quickly, and (2) postdamage information is usually unavailable. To address these drawbacks, we propose a new damage detection and localization approach based on stochastic time series analysis. Specifically, damage sensitive features, which are extracted from vibration signals, follow different distributions before and after a damage event. Hence, we use optimal change-point detection theory to find the time of damage occurrence. Because existing change-point detectors require the postdamage feature distribution, which is unavailable in SHM, we propose a maximum likelihood method for learning the distribution parameters from the time-series data. The proposed damage detection using estimated parameters achieves optimal performance. Also, we utilize the detection results to find damage location without any further computation. Validation results show highly accurate damage identification in American Society of Civil Engineers benchmark structures and two shake table experiments.

Distributed long gauge strain sensing technology has solved the problem of difficult identification of local damage in traditional ‘point’ monitoring, and has received extensive attention in the field of structural damage identification. Owing to the inevitable presence of measurement noise and environmental factors in the macro strain response measurement, a single damage index has also underlined some drawbacks generally arising when multiple damages occur, or errors affect the identified dynamic properties of the systems. To address these challenges, this paper proposes a data fusion method based on the Dempster–Shafer evidence theory, relying on distributed strain sensing technology. The identification results of the modal macro strain-based index and quasi-static macro strain energy-based damage index are fused to make a comprehensive decision on structural damage location. Damage identification studies are conducted on different types of structures under impact loads and random wind loads to verify the effectiveness and accuracy of the proposed data fusion method in the case of single and multiple damage conditions. The results show that the proposed data fusion method can accurately identify the damage location and effectively reduce misjudgment on undamaged locations; it has potential application value in practical structural health monitoring.

Structural health monitoring has been increasingly used due to the advances in sensing technology and data analysis, facilitating the shift from time-based to condition-based maintenance. This work is part of the efforts which have applied structural health monitoring to the Sydney Harbour Bridge – one of Australia’s iconic structures. It combines dimensionality reduction and pattern recognition techniques to accurately and efficiently distinguish faulty components from well-functioning ones. Specifically, random projection is used for dimensionality reduction on the vibration feature data. Then, healthy and damaged patterns of bridge components are learned in the lower dimensional projected space using supervised and unsupervised machine learning methods, namely, support vector machine and one-class support vector machine. The experimental results using data from a laboratory-based building structure and the Sydney Harbour Bridge showed high feasibility of applying machine learning techniques to dimensionality reduction and damage detection in structural health monitoring. Random projection combined with support vector machine significantly reduces the computational time while maintaining the detection accuracy. The proposed method also outperformed popular dimensionality reduction techniques. The computational time of the method using random projection can be more than 200 times faster than that without using dimensionality reduction while still achieving similar detection accuracy.

A generalized flexibility matrix based approach for structural damage detection

This article suggests the process of structural health monitoring (SHM) in the context of a nonstatistical damage detection paradigm. We particularly focus on applying the theory of possibility to the damage detection problem. The basic idea behind the proposed approach is that the application of possibility theory does not require probabilistic knowledge or assumptions on the damage feature and thus encompasses aleatoric and epistemic types of uncertainties. The approach is not damage feature dependent and thus is generic for use in many SHM systems. Additionally, two new damage metrics are introduced. These metrics extract information concerning damage evidence from observations performed at unknown health states of structures. Damage detection with the aid of the proposed approach is demonstrated by means of a case study.

A wavelet-based approach is proposed for structural damage detection and health monitoring. Characteristics of representative vibration signals under the wavelet transformation are examined. The methodology is then applied to simulation data generated from a simple structural model subjected to a harmonic excitation. The model consists of multiple breakable springs, some of which may suffer irreversible damage when the response exceeds a threshold value or the number of cycles of motion is accumulated beyond their fatigue life. In cases of either abrupt or accumulative damages, occurrence of damage and the moment when it occurs can be clearly determined in the details of the wavelet decomposition of these data. Similar results are observed for the real acceleration data of the seismic response recorded on the roof of a building during the 1971 San Fernando earthquake. Effects of noise intensity and damage severity are investigated and presented by a detectability map. Results show the great promise of the wavelet approach for damage detection and structural health monitoring.

Structural health monitoring (SHM) of bridges and, in particular, the development of efficient and reliable damage detection methodologies have been given a huge attention over the last three decades, by both the scientific community and bridge owners. Damage detection techniques involve the comparison of the structure’s response obtained in the monitoring stage with that of a baseline condition. The damage detection process generally involves five increasing knowledge steps ranging between, in a first stage, signalizing the occurrence of damage and, in a last stage, assessing its effects on the safety and lifecycle of the bridge. Therefore, detailed assessment and characterization of the in-service behaviour of the structure is paramount for successfully completing the damage detection process. With the aim of applying sound damage detection approaches based on the quasi-static component of moving-load responses in concrete bridges, this paper is devoted to the analysis of the in-service behaviour of these structures under moving-loads. A reduced-scale laboratorial model of a reinforced and prestressed concrete frame, subjected to a moving-load, was chosen as the case study. Although the laboratorial structure is not a scaled down model of a bridge, the authors believe that its response is representative of the behaviour of typical full-scale frame concrete bridges. Furthermore, it is noteworthy that smeared cracking might be observed along the frame’s beam, which is typical in concrete bridges. The structure under analysis was subjected to an extensive set of experimental tests, which comprised the observation of the structural response during the passage of a moving point-load, both forward and backwards, for different loads and under different prestress force scenarios (Cavadas & Figueiras 2014). That study revealed that, even though the pre-existing cracks remain entirely under compression (thanks to the applied prestressing forces) during the passage of the moving load, the structure exhibits a non-linear behaviour. This behaviour has important implications in the context of the application of damage detection techniques to cracked beam structures. On the one hand, it implies that the experimental influence lines become dependent on the direction of moving loads. This fact was evident in the experimental tests, where different influence lines for displacements, rotations and strains were obtained for forward and backward movement directions. On the other hand, the non-linear behaviour also implies that the beam’s deformations due to a constant point load will change if the prestressing force is modified. In this context, the main purpose of the present paper consists in the discussion and justification of the nonlinear behaviour observed in the laboratorial model. For that purpose, detailed finite element analyses were conducted. Nonlinear constitutive models were adopted to simulate the cyclic behaviour of cracked reinforced concrete structures. Special attention was devoted to the bond-slip behaviour at the interface between the reinforcing bars and the surrounding concrete, and also to the cyclic relationship between the local crack width and the concrete stress normal to the crack. The nonlinearity of concrete in compression was not considered because the structure is subjected to reduced concrete compressive stresses. The supports deformability was also considered in the analyses. Finally, it could be demonstrated that the observed nonlinear response is due to two main reasons: (i) the hysteretic behaviour of cracked reinforced concrete elements, which occurs even when pre-existing cracks are entirely under compression; (ii) the supports deformability. Even though the results and discussions presented in this paper were obtained for a reduced scale laboratorial model, the phenomena under analysis can be found in real structures and therefore, this subject is relevant in the context of the application of damage detection techniques to real bridge structures.

Motivated by a need to reduce energy consumption in wireless sensors for vibration-based structural health monitoring (SHM) associated with data acquisition and transmission, this paper puts forth a novel approach for undertaking operational modal analysis (OMA) and damage localization relying on compressed vibrations measurements sampled at rates well below the Nyquist rate. Specifically, non-uniform deterministic sub-Nyquist multi-coset sampling of response acceleration signals in white noise excited linear structures is considered in conjunction with a power spectrum blind sampling/estimation technique which retrieves/samples the power spectral density matrix from arrays of sensors directly from the sub-Nyquist measurements (i.e., in the compressed domain) without signal reconstruction in the time-domain and without posing any signal sparsity conditions. The frequency domain decomposition algorithm is then applied to the power spectral density matrix to extract natural frequencies and mode shapes as a standard OMA step. Further, the modal strain energy index (MSEI) is considered for damage localization based on the mode shapes extracted directly from the compressed measurements. The effectiveness and accuracy of the proposed approach is numerically assessed by considering simulated vibration data pertaining to a white-noise excited simply supported beam in healthy and in 3 damaged states, contaminated with Gaussian white noise. Good accuracy is achieved in estimating mode shapes (quantified in terms of the modal assurance criterion) and natural frequencies from an array of 15 multi-coset devices sampling at a 70% slower than the Nyquist frequency rate for SNRs as low as 10db. Damage localization of equal level/quality is also achieved by the MSEI applied to mode shapes derived from noisy sub-Nyquist (70% compression) and Nyquist measurements for all damaged states considered. Overall, the furnished numerical results demonstrate that the herein considered sub-Nyquist sampling and multi-sensor power spectral density estimation techniques coupled with standard OMA and damage detection approaches can achieve effective SHM from significantly fewer noisy acceleration measurements.

PurposeOne of the paramount challenges in the field of structural health monitoring (SHM) is the precise assessment of structural damage. Recently, many methods have been proposed to diagnose damage in structures using time series data. Nevertheless, these techniques are restricted in their ability to capture the bidirectional temporal and spatial relationships among the data. To remedy these shortcomings, this study proposes an advanced deep learning (DL) technique that integrates one-dimensional convolutional neural networks (1DCNN) with bidirectional long short-term memory (BiLSTM) networks, termed BiLSTM-1DCNN.Design/methodology/approachThis study proposes an advanced deep learning (DL) technique that integrates one-dimensional convolutional neural networks (1DCNN) with bidirectional long short-term memory (BiLSTM) networks, termed BiLSTM-1DCNN. The novelty of BiLSTM-1DCNN comes in its effective fusion of 1DCNN’s local feature extraction and BiLSTM’s capacity to capture long-term information from both directions (forward and backward).FindingsValidation using a dataset from the Nam O steel truss bridge demonstrates that the proposed method surpasses conventional models, including 1DCNN (87.9%), LSTM (72.5%), BiLSTM (76.1%) and LSTM-1DCNN (91.1%), achieving an accuracy rate of 96.4%. This approach provides a significant advancement in SHM, paving the way for more accurate and reliable damage detection.Research limitations/implicationsThe study’s limitations include potential overfitting due to the model’s complexity and the high computational resources required. The BiLSTM-1DCNN model’s performance is also influenced by the quality and quantity of the training data as well as the specific characteristics of the dataset. Additionally, while the model performs well in detecting damage in the Nam O steel truss bridge, its generalizability to other types of structures and environmental conditions remains untested.Practical implicationsThe BiLSTM-1DCNN model offers significant practical benefits for structural health monitoring by enhancing damage detection accuracy. Its ability to effectively capture both local features and long-term temporal dependencies makes it a valuable tool for real-time monitoring and early detection of structural issues. This can lead to more timely maintenance actions, potentially reducing repair costs and improving safety. The model’s high accuracy in detecting damage can also support infrastructure management decisions, contributing to better resource allocation and enhanced overall safety of critical structures.Social implicationsThe BiLSTM-1DCNN model’s advanced damage detection capabilities have positive social implications, such as improving the safety and reliability of public infrastructure like bridges. Enhanced structural health monitoring can reduce the risk of catastrophic failures, protecting lives and reducing accidents. By providing more accurate and timely assessments, the model can also lead to more efficient use of maintenance and repairs, ensuring that resources are allocated effectively.Originality/valueThe originality of the BiLSTM-1DCNN model lies in its innovative integration of one-dimensional convolutional neural networks (1DCNN) with bidirectional long short-term memory (BiLSTM) networks. This unique combination leverages 1DCNN’s strength in local feature extraction and BiLSTM’s capacity to capture temporal dependencies in both directions, addressing the limitations of existing methods. The model’s high accuracy in detecting structural damage, as demonstrated by its performance on the Nam O steel truss bridge dataset, highlights its value in advancing structural health monitoring technologies and offers a significant improvement over traditional approaches.

The propagation characteristics of Lamb waves in stiffened carbon fiber polymer matrix composite panels are investigated. In particular, the group velocities of S0 and A0 modes are calculated, the attenuation of Lamb waves induced by stiffeners and delaminations are evaluated, and the angular dependence of group velocity is subsequently examined. A two step damage detection and characterization approach, which uses both pulse-echo and pitchcatch active sensing schemes, is presented for the identification of delaminations in the stiffened composite panels. The matching pursuit decomposition algorithm is used for the representation of the sensor signal in the time-frequency domain. The delaminations are detected by identifying the converted Lamb wave modes in the time-frequency domain. The results confirm that the proposed approach is able to effectively identify the presence of seeded delaminations in complex composite structures.

Response surface methodology for damage detection using frequency and mode shape

SUMMARY In this study, a data-driven approach for structural seismic damage detection and localization in a multiple-degree-of-freedom shear-type building structure subjected to strong ground motion is presented. The proposed method is based on the joint implementation of time–frequency analysis and fractal dimension (FD) characteristics. The approximate analytical wavelet transform is first used to obtain the time–frequency feature (TFF) of the transient response at the measured story. The TFF is defined as the real part of the wavelet coefficients. Next, the box-counting method is used to acquire the FD of the TFF within the fundamental frequency band. It is verified that the proposed FDs at all stories of the linear system are identical, whereas the FDs at the stories with nonlinearities will be different from those at the stories with linearity. Therefore, the nonlinearity of the structure caused by strong ground motion can be detected and localized by comparing the FDs at the measured stories. A numerical simulation on a 10-story shear-type building was conducted. The simulation results indicate that the aforementioned approach is capable of detecting and localizing single-location or multiple-location seismic damages in shear-type building structures under various seismic excitations and is robust to measurement noise. Finally, the sensor placement for this approach, the effect of damping ratio, and choice of interstory drifts or accelerations as the measured signals are investigated and discussed. Copyright © 2012 John Wiley & Sons, Ltd.

This thesis is devoted to present some novel techniques in Structural Health Monitoring (SHM). SHM is a developing field that tries to monitor structures to make sure that they remain in their desired condition to avoid any catastrophe. SHM includes different levels from damage detection area to prognosis field. This work is dedicated to the first level, which might be considered the main and most important level. New techniques presented in this work are based on different statistical and signal processing methods such as Principal Component Analysis and its robust counterpart, Wavelet Transform, Fuzzy similarity, Andrew plots, etc. These techniques are applied on the propagated waves that are activated and captured in the structure using appropriate transducers. Piezoceramic (PZT) devices are chosen in this work to capture the signals due to their special characteristics such as high performance, low energy consumption and reasonable price. To guarantee the efficiency of the suggested techniques, they are tested on different laboratory and real scale test benchmarks, such as aluminum and composite plates, fuselage, wing skeleton, tube, etc. Because of the variety of tested benchmarks, this thesis is called damage detection in smart structures. This variety may promise the ability and capability of the proposed methods on different fields such as aerospace and gas/oil industry. In addition to the normal laboratory conditions, it is shown in this work that environmental changes can affect the performance of the damage detection and wave propagation significantly. As such, there is a vital need to consider their effect. In this work, temperature change is chosen as it is one of the main environmental fluctuation factors. To scrutinize its effect on damage detection, first, the effect of temperature is considered on wave propagation and then all the proposed methods are tested to check whether they are sensitive to temperature change or not. Finally, a temperature compensation method is applied to ensure that the proposed methods are stable and robust even when structures are subjected to variant environmental conditions. La presente tesis doctoral se dedica a la exploración y presentación de técnicas novedosas para la Monitorización y detección de defectos en estructuras (Structural Health Monitoring -SHM-) SHM es un campo actualmente en desarrollo que pretende asegurarse que las estructuras permanecen en su condición deseada para evitar cualquier catástrofe. En SHM se presentan diferentes niveles de diagnóstico, Este trabajo se concentra en el primer nivel, que se considera el más importante, la detección de los defectos. Las nuevas técnicas presentadas en esta tesis se basan en diferentes métodos estadísticos y de procesamiento de señales tales como el Análisis de Componentes Princpales (PCA) y sus variaciones robustas, Transformada wavelets, lógica difusa, gráficas de Andrew, etc. Estas técnicas de aplican sobre las ondas de vibración que se generan y se miden en la estructura utilizando trasductores apropiados. Dispositivos piezocerámicos (PZT's) se han escogido para este trabajo ya que presentan características especiales tales como: alto rendimiento, bajo consumo de energia y bajo costo. Para garantizar la eficacia de la metodología propuesta,se ha validado en diferentes laboratorios y estructuras a escala real: placas de aluminio y de material compuesto, fuselage de un avión, revestimiento del ala de un avóin, tubería, etc. Debido a la gran variedad de estructuras utilizadas, su aplicación en la industria aeroespacial y/o petrolera es prometedora. Por otra parte, los cambios ambientales pueden afectar al rendimiento de la detección de daños y propagación de la onda significativamente . En este trabajo , se estudia el efecto de las variaciones de temperatura ya que es uno de los principales factores de fluctuación del medio ambiente . Para examinar su efecto en la detección de daños, en primer lugar, todos los métodos propuestos se prueban para comprobar si son sensibles a los cambios de temperatura o no. Finalmente , se aplica un método de compensación de temperatura para garantizar que los métodos propuestos son estables y robustos incluso cuando las estructuras se someten a condiciones ambientales variantes