Structural Health Monitoring

Structural health monitoring systems are based on suitable sensor techniques allowing online and offline supervision of technical systems. The quantification of sensors/measurement devices is a key issue for qualifying their effectiveness and efficiency and therefore to ensure safe operations. Probability of detection serves as a performance measure for quantifying the reliability of conventional nondestructive testing procedures taking into account statistical variability of sensor-based measurements. For vibration-based supervision approaches and fault detection and isolation methods, the probability of detection approach cannot be applied similarly. This results mainly from the complexity of the dynamical behavior of systems monitored in relation to faults, sensors position (observability), and the related feature extraction or monitoring task. In this contribution, probability of detection evaluation of vibration-based fault detection of elastic mechanical structures to be monitored is developed. Beside a principal discussion of the problem serving as introduction, an example using different sensor types in combination with mechanical modifications of an elastic beam is presented. The a_90/95-criteria representing probability of 90% detection at a confidence level of 95% is examined to the measurements and related outcomes. Based on the analysis of a suitably chosen feature (like eigenfrequency or band power) and dependent on the mechanical modes considered, the efficiency and deficiency of the different combinations are shown. Based on the proposed approach, a new insight into the usefulness of different sensor type and fault position combination becomes possible. To improve the detection quality, suitable assumptions in combination with sensor/information fusion are applied to feature-based analysis as detection task using vibration measurements. In addition, based on an experimental evaluation, it can be concluded that a suitable fault-feature probability of detection analysis can be successfully implemented as a new reliable measure for vibration-based fault detection and isolation approaches. Furthermore, decision fusion as the combination of different measurements will allow the improvement of results. Dependent on noise analysis, a trade-off between flaw size detection and probability of falsely characterizing a fault with a_90/95 reliability level can be attained.

Concrete structures in service are often subjected to environmental/operational temperature effects, which change their inherent properties and also inflict a challenge to their extrinsic monitoring systems. Recently, piezoelectric lead zirconate titanate (PZT)-based electromechanical admittance technique has been increasingly growing into an effective tool for concrete structural health monitoring; however, uncertainty in the changes of monitoring signals induced by temperature impact on concrete/PZT sensor would inevitably cause interference to structural damage detection, which adversely hinder its application from laboratory to engineering practice. This article, aiming at exploring the temperature effect on the electromechanical admittance–based concrete damage evaluation, primarily covered a series of theoretical/numerical analysis with rigorously experimental verifications. Three aspects of comparative studies were performed in theoretical/numerical analysis: (1) thermal-dependent parameters were inclusively evaluated in contribution to the electromechanical admittance characteristics via PZT-structure interaction models; (2) three-dimensional finite element analysis in multi-physics coupled field was employed to qualitatively assess the singular temperature effect on the electromechanical admittance behaviors of free-vibrated PZT, surface-bonded PZT/inside-embedded PZT coupled healthy concrete cubes; and (3) depending on the modeling of surface-bonded PZT-/inside-embedded PZT-cracked concrete cube, thermal effect on damage evaluation was addressed via quantification on the electromechanical admittance variations. In the experimental study, rigorous validation tests were carried out on a group of lab-scale concrete cubes, where surface-bonded PZT/inside-embedded PZT transducers were simultaneously employed for electromechanical admittance monitoring in view of thermal difference between concrete surface and its inner part. Correlation coefficient deviation value-based effective frequency shifts algorithm was also employed to compensate the temperature effect. Moreover, temperature effect was further testified on the monitoring of a full-scale shield-tunnel segment structure. Experimental results indicated that temperature triggered different behaviors of electromechanical admittance signatures for surface-bonded PZT/inside-embedded PZT transducers and contaminated the electromechanical admittance responses for damage detection. Structural damage severity level can be disadvantageously amplified by temperature increment even if under the same damage scenarios.

The importance of wireless sensor networks in structural health monitoring is unceasingly growing, because of the increasing demand for both safety and security in the cities. The speedy growth of wireless technologies has considerably developed the progress of structural monitoring systems with the combination of wireless sensor network technology. Wireless sensor network–based structural health monitoring system introduces a novel technology with compelling advantages in comparison to traditional wired system, which has the benefits of reducing installation and maintenance costs of structural health monitoring systems. However, structural health monitoring has brought an additional complex challenges in network design to wireless sensor networks. This article presents a contemporary review of collective experience the researchers have gained from the application of wireless sensor networks for structural health monitoring. Technologies of wired and wireless sensor systems are investigated along with wireless sensor node architecture, functionality, communication technologies, and its popular operating systems. Then, comprehensive summaries for the state-of-the-art academic and commercial wireless platform technologies used in laboratory testbeds and field test deployments for structural health monitoring applications are reviewed and tabulated. Following that, classification taxonomy of the key challenges associated with wireless sensor networks for structural health monitoring to assist the researchers in understanding the obstacles and the suitability of implementing wireless technology for structural health monitoring applications are deeply discussed with available research efforts in order to overcome these challenges. Finally, open research issues in wireless sensor networks for structural health monitoring are explored.

Random-vibration-based statistical time series structural health monitoring methods utilize small-scale, compact, and data-based, time series stochastic representations of the structural dynamics for damage diagnosis. In this study, a comprehensive and critical assessment of the diagnostic performance of five prominent response-only methods is presented based on incipient, ‘minor’ to ‘mild’, damages on a lab-scale wind turbine jacket structure. Statistically reliable damage detection and identification results are obtained via a ‘rotation’ procedure resulting into thousands of test cases, with the performance analysed in terms of receiver operating characteristic curves and confusion matrices. The results indicate not only challenging of the methods’ capabilities, but also the achievement of good to excellent performance for the ‘minor’ to ‘mild’ damages, respectively, with the model parameter–based method offering the best performance. In addition, the use of a vibration signal measured via a laser vibrometer leads to slightly improved detection performance over that obtained via a classical accelerometer.

Cavities with different geometries represent the internal volumes of various engineering applications such as cabins of passenger cars, fuselages and wings of aircraft, and internal compartments of wind turbine blades. Transmissibility of acoustic excitation to and from these cavities is affected by material and cross-sectional properties of the structural cavity, as well as potential damage incurred. A new structural damage detection methodology that relies on the detectability of the changes in acoustic transmissibility across the boundaries of structural cavities is proposed. The methodology is described with a specific focus on the passive damage detection approach applied to cavity internal acoustic pressure responses under external flow-induced acoustic excitations. The approach is realized through a test plan that considers a wind turbine blade section subject to various damage types, severity levels, and locations, as well as wind speeds tested in a subsonic wind tunnel. A number of statistics-based metrics, including power spectral density estimates, band power differences from a known baseline, and the sum of absolute difference, were used to detect damage. The results obtained from the test campaign indicated that the passive acoustic damage detection approach was able to detect all considered hole-type damages as small as 0.32 cm in diameter and crack-type damages 1.27 cm in length. In general, the ability to distinguish damage from the baseline state improved as the damage increased in severity. Damage type, damage location, and flow speed influenced the ability to detect damage, but were not significant enough to prevent detection. This article serves as an overall proof of concept of the passive-based damage detection approach using flow-induced acoustic excitations on structural cavities of a wind turbine blade. The laboratory-scale results reveal that acoustic-based monitoring has great potential to be used as a new structural health monitoring technique for utility-scale wind turbine blades.

Conductive nanocomposites have been explored extensively for structural health monitoring (SHM) due to their self-sensing nature via the piezoresistive effect. Combined with a non-invasive conductivity imaging modality such as electrical impedance tomography (EIT), piezoresistivity is a powerful tool for SHM. To date, however, the combination of the piezoresistive effect and EIT has been limited to just damage detection. From a SHM perspective, it may be more beneficial to pre-emptively predict failure before it occurs. To that end, we propose a novel methodology for failure prediction in nanocomposites using piezoresistive inversion. Our approach makes use of a genetic algorithm (GA) to determine the mechanical state of the structure using conductivity changes observed via EIT. First, a rectangular nanocomposite specimen with a central hole is manufactured. Second, the specimen is loaded in tension to induce stress concentrations near the hole. Third, EIT is used to image the resulting stress concentration-induced conductivity changes near the hole. Fourth, GA-enabled piezoresistive inversion is implemented to determine the underlying displacements from the observed conductivity changes. The strains are then determined from kinematic relations and the stresses from constitutive relations. Lastly, a failure criterion is used to predict failure. By validating our results with finite element analysis and digital image correlation, we demonstrate that the proposed approach can accurately predict the onset of failure and therefore enable unprecedented SHM capabilities in piezoresistive nanocomposites.

In previous works, the authors have shown the feasibility of using classical and robust principal component analysis for damage detection on structures when ultrasonic guided waves are used. It has also been demonstrated that robust principal component analysis presents a higher probability of detection accuracy when data are corrupted. In the present work, a robust principal component analysis orthogonal distance is proposed as a new feature for damage detection strategy based on ultrasonic guided waves on structures subjected to uniform temperature changes. The effect of this temperature fluctuation on the signal propagation and also in the new feature is analyzed. Temperature compensation is applied to mitigate the effect of temperature changes on the reliability of the damage detection methodology. The proposed feature and damage detection strategy that considers these effects are tested on two structures: a laboratory scale composite plate and a large-scale complex composite that is representative of a component from an aerospace application. The promising result proves the ability of the new feature as a damage detection tool.

Anomaly recognition and early warning of monitoring data are of great significance in the field of modern dam safety management. Multidimensional least-squares regression model with the Pauta criterion is a well-known traditional method, but it is easy to misjudge the normal value and miss the outliers. Thereby, an online robust recognition and early warning model combining robust statistics and confidence interval is proposed to detect outliers. The threshold 3 S T + D is set based on the derived confidence interval D and the scale estimator S T (derived from the location M-estimator). Monitoring data obtained from a gravity dam and a rockfill dam were taken as examples to demonstrate the robust recognition and early warning model. The results show that the proposed method can effectively improve the reliability of anomaly recognition and early warnings, which is valuable in engineering applications.

In this article, a robust output-only real-time damage detection technique for multi-degree-of-freedom degrading systems using recursive canonical correlation analysis is presented. It has been observed that the impact of damage to a vibrating system gradually advances with time that sustains until the system degrades up to a considerable extent. Of significant interest is the effect of sudden damage in presence of continuous degradation in real-time, which is studied in the form of a sudden stiffness reduction in a separate floor. The proposed recursive canonical correlation analysis algorithm estimates the iterative update of eigenspace at each instant from the response data, thereby capturing the features of a time varying degrading structure in an online framework. Furthermore, recursive canonical correlation analysis algorithm is shown to reduce the computational cost by updating the eigenspace at each instant of time. This article explores newly developed recursive condition indicators: recursive Mahalanobis distance and recursive Itakura distance that elicit damage information from the eigenspace. In order to model degradation, simulations aimed at successfully capturing the behavior of the process in real-time becomes imperative. A general stochastic formulation of the coupled response-degradation problem accounting for the evolution of degradation is presented in the light of stiffness degradation problems. The evolution of time varying system responses is generated using a newly proposed Ito–Taylor expansion-based stochastic numerical integration formulation. Numerically simulated structural vibrating systems, namely, 2-degree-of-freedom base-isolated and 4-degree-of-freedom linear systems, have been used to check the performance of the recursive canonical correlation analysis method. The spatial damage detectability of the algorithm in real-time is explored through identifying crack location on a beam traversed by a vehicle. Finally, an experimental case study has been carried out to verify the robustness of the proposed algorithm. The identification results for both numerical and experimental cases demonstrate the efficacy of the proposed algorithm in identification of nonlinear and time varying behavior associated with degrading structural systems.

In this article, a framework of building monitoring is developed and transient and permanent variations of the fundamental period of vibration caused by both damage and repair interventions are investigated. The buildings of the University of Ferrara (Emilia-Romagna Region, Northern Italy), struck by the 2012 Emilia seismic sequence, were monitored using both temporary and permanent equipment: the first one to perform ambient vibration tests and the second one to implement permanent real-time monitoring for earthquake recording. Three on-demand ambient vibration tests were performed at each floor of the buildings: the first dataset was acquired a few months before the mainshock occurred on 20 May 2012 (M_L = 5.9, 6.8 km depth and 30 km epicentral distance); the second was acquired right after the end of the sequence, when the building showed slight damage (degree 1 according to the European Macroseismic Scale 98); finally, the third dataset was acquired in 2016 after the repair intervention. The data analysis clearly documented the permanent drop of the first vibration frequency as a symptom of the damage and its partial recovery that followed the repairs. The permanent real-time monitoring system, despite the fact that it was implemented using low-cost sensors, provided an insight into the intra-event frequency variation, allowing in turn a preliminary damage assessment.

This article presents an ensemble decision tree–based random forest regression methodology for remaining useful life prediction of spur gears subjected to pitting failure mode. The random forest regression methodology does not require an elaborate statistics background knowledge and has an inbuilt health indicator selection capability compared to other existing data-driven remaining useful life prediction approaches. A correlation coefficient parameter based on the residual vibration signal is used for monitoring and detecting the pitting progression in spur gears. The effectiveness of the correlation coefficient of the residual vibration signal is assessed over the other existing health indicators for pitting fault progression. To show the inbuilt best health indicator selection capability of the random forest regression model, initially, eight indicators (existing seven and the correlation coefficient of the residual vibration signal) were used for model training. In addition, the effect of fusing the vibration sensor data from multiple positions on the gearbox on prediction accuracy of the random forest regression model is also evaluated. The accuracy in the remaining useful life prediction is found to increase after fusing the correlation coefficient of the residual vibration signal based health indicator derived from the accelerometers located at multiple positions on the gearbox in comparison to data from a single accelerometer. Furthermore, the accuracy of the proposed methodology is tested and proven using five accelerated run-to-failure experimental data collected from the specially built test rig.

It is significant to perform looseness condition detection of viscoelastic sandwich structures to avoid serious accidents. Due to the multilayer characteristic of the viscoelastic sandwich structure, the vibration response signal of such structures is nonlinear and nonstationary. Furthermore, the looseness condition feature signal contained in the vibration response signal is very puny. Condition feature extraction has become a challenging task in the looseness condition detection of viscoelastic sandwich structures. Therefore, a novel method called dual-tree complex wavelet packet-based deep autoencoder network is proposed for this task. First, the vibration response signal of the viscoelastic sandwich structure is decomposed by dual-tree complex wavelet packet transform and the sub-band signals which contain rich energy are extracted. Then, the energies of the extracted sub-band signals are calculated to form a feature set. Finally, a deep autoencoder network is established to fuse the feature set, and the fused feature is viewed as the detection index to detect the looseness condition of the viscoelastic sandwich structure. The proposed method is applied to the connecting bolt looseness condition detection of the viscoelastic sandwich structure to validate its effectiveness. Compared with the detection method based on dual-tree complex wavelet packet transform and energy and the detection method based on dual-tree complex wavelet packet transform and permutation entropy, the results indicate that the effectiveness of the proposed method in this article is more superior to that of the other two methods.

Composites are prone to failure during operating conditions and that is why vast research studies have been carried out to develop in situ sensors and monitoring systems to avoid their catastrophic failure and repairing cost. The aim of this research article was to develop a flexible strain sensor wire for real-time monitoring and damage detection in the composites when subjected to operational loads. This flexible strain sensor wire was developed by depositing conductive silver (Ag) nanoparticles on the surface of nylon (Ny) yarn by electroless plating process to achieve smallest uniform coating film without jeopardizing the integrity of each material. The sensitivity of this Nylon/Ag strain sensor wire was calculated experimentally, and gauge factor was found to be in the range of 21–25. Then, the Nylon/Ag strain sensor wire was inserted into each composite specimen at different positions intentionally during fabrication depending upon the type of damage to detect. The specimens were subjected to tensile loading at a strain rate of 2 mm/min. Overall mechanical response of composite specimens and electrical response signal of the Nylon/Ag strain sensor wire showed good reproducibility in results; however, the Nylon/Ag sensor showed a specific change in resistance in each direction because of the respective position. The strain sensor wire designed not only monitored the change in the mechanical behavior of the specimen during the elongation and detected the strain deformation but also identified the type of damage, whether it was compressive or tensile. This sensor wire showed good potential as a flexible reinforcement in composite materials for in situ structural health monitoring applications and detection of damage initiation before it can become fatal.

Imaging a damage using the phase of wavefield “video” in the physical domain is developed and applied to a stiffened curved composite sandwich panel for visualization of a barely visible impact damage. The ultrasonic guided waves are generated by the thermoelastic effect induced by a Nd:YAG Q-switched pulse laser and then are captured point-by-point by a laser Doppler vibrometer. The wavefield “video” is reconstructed by measuring out-of-plane velocity on the smooth outside surface of the panel. Newly generated wavenumbers from the geometry/material discontinuities caused by either the stiffener or the impact damage can be detected by observing the change of wavenumber from the reconstructed wavefield “video.” The instantaneous wavenumber (i.e. the magnitude of instantaneous wavevector) derived via Riesz transform and its difference can be shown and highlighted using a proposed imaging condition, named as wavenumber index. The wavenumber index sums the wavenumber values followed by a wave energy threshold filter which is performed in the time domain. This is in contrast to other imaging conditions implemented in the frequency-wavenumber domain by the use of complex wavenumber filtering and wave mode decomposition. Since wavenumber index is the phase-based imaging technique instead of conventional intensity-based technique for the wavefield “video,” this technique is robust in that the impact damages located in the vicinity of geometry/material discontinuities can yield consistent damage image resolution with high sensitivity even for wave propagating from the direction across the stiffener. The barely visible impact damage of the composite structure becomes therefore “visible” with the proposed imaging technique.

A novel damage detection approach using only two sensors to detect the damage in beam bridges subjected to a moving vehicle is proposed in this article. In this approach, a moving mass is considered representing a vehicle moving across the bridge, and structural vibration responses at two locations are measured from a pair of sensors. A moving window is defined with a certain length determined by the sampling frequency and the fundamental frequency of the measured responses. The windowed pair time series extracted from these two measured responses are used to calculate the cross-correlation, which is used to define the local damage index. A simply supported beam bridge subjected to a moving mass is simulated to demonstrate the effectiveness and accuracy of the proposed approach. Numerical results indicate that the proposed approach can accurately identify the single and multiple damages using both displacement and acceleration responses, even when the responses are smeared with a significant noise. This indicates a good robustness to the noise effect. Experimental verifications on a laboratory beam bridge model demonstrate that the proposed approach can successfully identify the damage location using different selections of sensor pairs. Both the numerical and experimental results demonstrate that the new damage index is a good candidate for structural damage detection with very limited measurement information.

A novel nondestructive testing methodology is proposed for online monitoring health and for in situ determination of the fatigue life of a component. This method only requires the determination of current and voltage consumed by the driving motor during its operation. The efficacy of the proposed method is established by performing a series of bending fatigue experiments over a wide range of operating conditions with specimens made of SS304 and SS321 materials. Results are correlated with those obtained thermographically via an infrared camera. To further establish the utility of the approach, tests were conducted with Aluminum 7075-T6 material that experiences minimal temperature rise and renders the thermographic method inefficient. The results show that the proposed method is effective in monitoring the failure of the specimen accurately and reliably.

Guided waves allow for the efficient structural health monitoring of large structures using phased or distributed arrays of sensors. The sensitivity for specific defects can be improved by accounting for the angular scattering pattern. The scattering of the fundamental anti-symmetric guided wave mode (A₀ Lamb mode) at through-thickness and part-through crack-like defects was studied experimentally and from three-dimensional finite element simulations. Experimentally, the scattered field around manufactured notches of different depths and lengths in an aluminium plate was measured using a laser interferometer. The scattered field was extracted by taking the complex difference in the frequency domain between baseline measurement and measurements around the defect. Good agreement was found between measurements and three-dimensional finite element simulations, and the amplitude and directionality pattern of the scattered field can be predicted accurately. The angular directionality pattern of the scattered field depending on the direction of the incident wave relative to the crack-like defect orientation, depth and length relative to the wavelength was investigated. For short and part-thickness defects, the main scattering effect is a reduction of the (forward) wave propagating past the defect with very limited backscattered amplitude. Significant energy scattered back towards the incident wave direction was only found for perpendicular incidence on long and deep defects. Even for large defects, almost no energy is scattered in certain directions from the defect, possibly complicating defect detection. Based on the predicted amplitude and angular dependency of the scattered guided waves, the sensitivity for defect detection using localized and distributed structural health monitoring sensor array systems can be quantified.