Structural Health Monitoring Approach Research Articles

Pulsed terahertz spectroscopy combined with hybrid machine learning approaches for structural health monitoring of multilayer thermal barrier coatings.

Structural health monitoring of multilayer thermal barrier coatings (TBCs) is very vital to ensure the structural integrity and service performance of the hot-section components of the aero-engine. In this paper, we theoretically and numerically demonstrated that the terahertz time domain spectrum and the terahertz reflectance spectrum could be adopted to estimate the structure parameters, based on the finite difference time domain (FDTD) algorithm, 64 samples which were imported with three kinds of 64 sets structure parameters had been calculated to obtain the time domain and terahertz reflectance signals. To mimic the actual test signals, the original FDTD simulation signals were processed by adding the Gaussian white noise and wavelet noise reduction. To reduce the data dimension and improve the calculation efficiency during modeling, the principal component analysis (PCA) algorithm was adopted to reduce the dimensions of time-domain data and reflectance data. Finally, these data after multiple signal processing and PCA feature extraction were used to train the extreme learning machine (ELM), combining the genetic algorithm (GA) could optimize the PCA-ELM model and further improve the prediction performance of the hybrid model. Our proposed novel and efficient terahertz nondestructive technology combined with the hybrid machine learning approaches provides great potential applications on the multilayer TBCs structural integrity evaluation.

Implementation of PPP-GNSS measurement technology in the probabilistic SHM of bridge structures

The use of geodetic-grade GNSS (Global Navigation Satellite System) receivers is increasingly evolving in the Structural Health Monitoring (SHM) of infrastructure. In this sense, the Precise Point Positioning GNSS (PPP-GNSS) is becoming very popular since it requires only one GNSS receiver. This paper presents a novel SHM technique by integrating the PPP-GNSS measurement technology and an alternative probabilistic approach. Such an approach is based on the study of the stochastic behavior of displacements and their corresponding probability density functions. As a case of study, the El Carrizo Bridge is studied. Results validate the accuracy of the PPP-GNSS measurement technology, and the potential benefits of the alternative probabilistic SHM approach since risk can be extracted using a single GNSS receiver in terms of reliability index and probability of failure. Finally, based on the probabilistic evaluation, it is demonstrated that the El Carrizo Bridge presents an adequate behavior after its rehabilitation.

A methodological approach for structural health monitoring of mass-timber buildings under construction

Structural health monitoring (SHM) is a method used to evaluate the performance of new structural systems and critical infrastructure. With mass-timber building construction on the rise, SHM programs have emerged to document hygrothermal, static, and dynamic behavior of these structures. To most efficiently document behavior and provide recommendations to industry, it is key that the research community work collaboratively to create consistent data by using standardized approaches. This paper presents a methodological approach for monitoring mass-timber buildings during construction to address this need. The approach was validated over ten months with a mass-timber building under construction at Oregon State University.

A Fuzzy Krill Herd Approach for Structural Health Monitoring of Bridges using Operational Modal Analysis

Monitoring bridges is an important issue in the structural safety assessment. As a consequence of inaccessible sections of these structures, the utilization of nondestructive damage identification techniques seems to be vital for maintaining safety and integrity. In this study, a new hybrid Fuzzy Krill Herd approach is performed based on online responses to assess the global structural integrity of an in-service bridge. While the algorithms have been studied previously in the literature, their efficiency in structural health monitoring of a real bridge structure has yet to be investigated. The methodology is applied to two types of steel girders and concrete bridges in this manuscript. The finite element (FE) modeling is considered for further numerical investigation of dynamic characteristics and structural behavior of the bridges. To evaluate the efficiency of the proposed technique, a three-dimensional FE model and a developed simple two-dimensional girders models are simulated. The results indicate the capability of the fuzzy logic approach in obtaining accurate information in the presence of noisy input data or the data with missing values. Based on the results, increasing the number of the measurement modes and using the torsional modes lead to an accurate damage diagnosis process even in symmetric structures.

Acousto-ultrasonics-based health monitoring for nano-engineered composites using a dispersive graphene-networked sensing system

Sensing is a fundamental yet crucial part of a functional structural health monitoring system. Substantial research has been invested in developing new sensing techniques to enhance sensing efficiency and accuracy. Practical applications of structural health monitoring approaches to real engineering structures require strict criteria for the sensing system (e.g. weight, position, intrusion and endurance), which challenge existing sensing techniques. The boom in nanotechnology has offered promising solutions for the development of new sensing approaches. However, a bottleneck still exists when considering the density of sensors and surface-mounted modality of installation. In this study, graphene nanoparticles are dispersed into a glass fibre/epoxy composite to form a dispersive network sensing system. The piezoresistivity of the graphene-formed network changes locally as a result of the change of inter-nanoparticle distances which triggers the ‘tunnelling effect’ and drives the sensor to respond to propagating elastic waves. Due to the dense graphene network formed within the composite, only a small area is required, functioning as a single sensing element to capture ultrasonic waves. To validate such capability, passive acoustic emission tests and active guided ultrasonic wave tests are performed individually. The graphene-networked sensing system can precisely capture wave signals which contain effective features to identify impact spot or damage location. Integrating passive graphene-formed network and active lead zirconate titanate wafers can form a dense network, capable of fulfilling general structural health monitoring tasks.

Reliability of crack quantification via acousto-ultrasound active-sensing structural health monitoring using surface-mounted PZT actuators/sensors

This work presents the introduction and experimental investigation of an active-sensing acousto-ultrasound structural health monitoring approach for damage size quantification based on piezoelectric sensors/actuators mounted on multiple seemingly identical structural components. The objective of this work is to determine how reliable the damage diagnostics can be from one component to another similar (nominally identical) component using surface-mounted PZT (lead zirconate titanate) sensors/actuators, and also to evaluate how sensitive a sensor network configuration in terms of the number of sensors/actuators is with respect to its detection reliability. Extensive crack growth experiments on multiple identical coupons outfitted with the same sensor network configuration under cyclic loads were conducted to assess the damage quantification reliability from one coupon to another using the same diagnostic algorithm. The results of the study indicate that the crack size estimates obtained from the active-sensing structural health monitoring system can vary within the population of identical structural components (coupons), but the difference in quantifying damage among coupons decreases with the increase in the number of sensors and actuators used, that is, wave propagation paths. Furthermore, it is shown that the diagnostic results in terms of damage quantification converge with the increase in the number of sensors. The results of the study indicate that the diagnostic approach using a multi-path sensor network can reduce the damage quantification error from one component to another within a “hotspot” configuration (damage location is known or suspected a priori). Finally, the results of this study indicate that the more wave propagation paths used in the diagnostic active-sensing algorithm, the more reliable the damage quantification results are, provided that the same sensor network is used and installed at nominally identical locations for all coupons.

Application of short‐time matrix pencil method for high‐frequency damage detection in structural system

Structural health monitoring (SHM) has been becoming a key element of monitoring and maintenance in modern and smart cities with critical infrastructures. Within the past decade, SHM approaches are widely used in bridges, roads, skyscrapers, critical facilities, and buildings to enhance the public safety and cost efficiencies of infrastructure maintenance under extreme events, such as earthquakes. Several advanced wave-based methods, such as sliding discrete Fourier transform, discrete wavelet transform (DWT), and Hilbert transform were broadly used to detect different failure modes of systems. One of the most important deficiencies associated with the state-of-the-art approaches is their high level of sensitivity to the sampling frequency, measurement noise, and damage-induced pulse parameters. To fill this technical gap, for the first time, this paper proposes a high-frequency damage detection approach based on the short-time matrix pencil method (STMPM) that has been used in electrical engineering lately. The advantage of STMPM compared with the similar approaches is using the complex frequencies that are not used before in damage detection applications. In this study, pounding between two adjacent buildings is considered as a case study of high-frequency damage scenario. The proposed STMPM-based detection technique is compared with DWT with commonly used mother wavelets. According to the simulation results, the STMPM-based detection is more robust compared with DWT-based techniques and has lower sensitivity to the noise parameters and sampling rate.

An acoustic emission based structural health monitoring approach to damage development in solid railway axles

The in-service safety of railway axles is a very important engineering challenge, as it has a large impact not only from the economic point of view of the railway operator, but it has cascading effects on supply chains, loss of work productivity, and, in the most serious cases, loss of life. It is, therefore, vital that the structural integrity of such components is known, during their lifecycle, with the highest possible accuracy via precise modelling, reliable inspections and, more recently but still at research level, effective condition monitoring.With a focus on solid freight axles, the research investigates the applicability of Acoustic Emission as a structural health monitoring approach for determining the in-service condition of a full-scale axle. A fatigue crack propagation test is carried out in the lab subjecting the axle to many repetitions of a block load sequence defined from real service measurements. Acoustic Emission data are continuously recorded during the test, whilst crack size is periodically measured by conventional non-destructive techniques.Eventually, a first-approximation correlation is highlighted between Acoustic Emission data, post-processed by a machine-learning algorithm, and crack propagation ones.

Fatigue damage monitoring of reinforced concrete frames using wavelet transform energy of PZT-based admittance signals

Abstract The paper presents a novel structural health monitoring approach based on the wavelet energy of admittance signals to detect, localize and estimate the severity of the damages caused by fatigue in reinforced concrete (RC) frames. Admittance signatures are acquired in user-defined frequency ranges experimentally using six piezo-cement composite disks (embedded at different locations in RC specimens) at various stages during fatigue testing of the frame on a shake table. For purpose of estimating wavelet energies, discrete wavelet transformation (DWT), continuous wavelet transformation (CWT) and power spectral density (PSD) analysis are applied on real-admittance signals (conductance) in frequency domains. Mathematical models based on global dynamic and wavelet energy approach are developed and validated in order to estimate residual life of RC frames. Overall, the experimental results verify the superior performance of the DWT based optimum methodology in enabling full-fledged real-time damage prognosis of RC structures under low-strain and high-cycle fatigue.

On the vulnerability of data-driven structural health monitoring models to adversarial attack

Many approaches at the forefront of structural health monitoring rely on cutting-edge techniques from the field of machine learning. Recently, much interest has been directed towards the study of so-called adversarial examples; deliberate input perturbations that deceive machine learning models while remaining semantically identical. This article demonstrates that data-driven approaches to structural health monitoring are vulnerable to attacks of this kind. In the perfect information or ‘white-box’ scenario, a transformation is found that maps every example in the Los Alamos National Laboratory three-storey structure dataset to an adversarial example. Also presented is an adversarial threat model specific to structural health monitoring. The threat model is proposed with a view to motivate discussion into ways in which structural health monitoring approaches might be made more robust to the threat of adversarial attack.

Dynamic response reconstruction for structural health monitoring using densely connected convolutional networks

This article proposes a novel dynamic response reconstruction approach for structural health monitoring using densely connected convolutional networks. Skip connection and dense block techniques are carefully applied in the designed network architecture, which greatly facilitates the information flow, and increases the training efficiency and accuracy of feature extraction and propagation with fewer parameters in the network. Sub-pixel shuffling and dropout techniques are used in the designed network and applied to reduce the computational demand and improve training efficiency. The network is trained in a supervised manner, where the input and output are the measurements of the available channels at response available locations and desired channels at response unavailable locations. The proposed densely connected convolutional networks automatically extract the high-level features of the input data and construct the complicated nonlinear relationship between the responses of available and desired locations. Experimental studies are conducted using the measured acceleration responses from Guangzhou New Television Tower to investigate the effects of the locations of available responses, the numbers of available and unavailable channels, and measurement noise. The results demonstrate that the proposed approach can accurately reconstruct the responses in both time and frequency domains with strong noise immunity. The reconstructed response is further used for modal identification to demonstrate the usability and accuracy of the reconstructed responses. The applicability of the proposed approach for structural health monitoring is further proved by the highly consistent modal parameters identified from the reconstructed and true responses.

An integrated investigative approach in health monitoring of masonry arch bridges using GPR and InSAR technologies

This paper provides an overview of the existing health monitoring and assessment methods for masonry arch bridges. In addition, a novel “integrated” holistic non-destructive approach for structural monitoring of bridges using ground-based non-destructive testing (NDT) and the satellite remote sensing techniques is presented. The first part of the paper reports a review of masonry arch bridges and the main issues in terms of structural behaviour and functionality as well as the main assessment methods to identify structural integrity-related issues. A new surveying methodology is proposed based on the integration of multi-source, multi-scale and multi-temporal information collected using the Ground Penetrating Radar (GPR – 200, 600 and 2000 MHz central-frequency antennas) and the Interferometric Synthetic Aperture Radar (InSAR – C-band SAR sensors) techniques. A case study (the “Old Bridge” at Aylesford, Kent, UK – a 13th century bridge) is presented demonstrating the effectiveness of the proposed method in the assessment of masonry arch bridges. GPR has proven essential at providing structural detailing in terms of subsurface geometry of the superstructure as well as the exact positioning of the structural ties. InSAR has identified measures of structural displacements caused by the seasonal variation of the water level in the river and the river bed soil expansions. The above process forms the basis for the “integrated” holistic structural health monitoring approach proposed by this paper.

An acoustic emission-based approach to structural health monitoring of pre-stressed concrete railway sleepers

Railway sleepers represent an essential element of the track; indeed, their structural integrity is closely related to important technical and safety issues. Today, periodical visual inspections are the only method applied to check the status of sleepers but are limited to visible surfaces, whereas the early detection of in-service cracks in the whole volume of sleepers would provide great advantages in terms of maintenance and management. The aim of the paper is to propose an acoustic emission (AE)-based structural health monitoring (SHM) approach that is able to detect the initiation and propagation of cracks in in-service pre-stressed concrete sleepers. The investigation is carried out in the laboratory, comparing the results obtained by acoustic emission monitoring and digital image correlation when subjecting pre-stressed concrete sleepers, taken from production, to both static and cyclic loads. The main points dealt with in the paper include the sensitivity of acoustic emission to detect damage development and the signal processing approach needed for defining effective damage indexes. Given the encouraging results, the paper is the first step in developing an affordable monitoring system, to be embedded into sleepers, that is able to be part of a complete track monitoring system.

Data-Driven Structural Health Monitoring Approach Using Guided Lamb Wave Responses

AbstractIn this paper, a data-driven structural health monitoring (SHM) approach is proposed to conduct in situ evaluation of the structural health state, i.e., damage location and extent, in which...

Bolt-looseness detection by a new percussion-based method using multifractal analysis and gradient boosting decision tree

Among various building blocks, bolted connections are the most widely used components, which can be employed to hold the integrity of entire structures. Looseness detection of bolted connections has been an attractive topic in multiple fields including aerospace and mechanical engineering, since loose-induced bolt failures may lead to costly disasters. Recently, several structural health monitoring methods have been applied to detect bolt looseness; however, they are often impeded in practical use due to the requirement for constant sensor–structure interaction. Thus, the potential of the percussion-based method in detecting bolt looseness has been noticed. In this article, considering the drawbacks existing in prior investigations (e.g. manual feature selection), a new percussion-based method was proposed to inspect bolt looseness. Based on the multifractal analysis and the joint mutual information maximization method, the feature sets of percussion-induced sound signals were selected automatically, which effectively avoided highly experienced personnel for manual feature selection. Subsequently, after feeding extracted feature sets into a gradient boosting decision tree model, we trained a classifier to achieve the identification of bolt looseness. Compared to current percussion-based methods for bolt-loosening detection, the method we proposed in this article has higher accuracy, which is proven by experimental results. Finally, as a rapid and non-invasive structural health monitoring approach, our method can be applied to detect damages in other structures and thus guides future investigations.

Damage detection in a fixed-fixed beam using natural frequency changes

This paper presents a damage detection technique which uses change in Frequency Response Function (FRF) and Modal Strain Energy (MSE) for damage detection in beam and truss like structures. It’s a robust approach for Structural Health Monitoring (SHM) using natural frequency for structural damage assessment. This technique couples FRF with Iterative Modal Strain Energy (IMSE) method to accurately localize and quantify the damage with only few frequencies, overcoming the limitation of mode shape based damage detection methods which requires mode shapes at full coordinates but damage is either over/underestimated. In the current approach, a damage detection algorithm is developed and validated by conducting numerical studies for a Fixed-Fixed beam, both noise-free and noise-effected cases are simulated. The numerical studies reveal that proposed algorithm is capable of identifying, localizing and quantifying the damage accurately at reduce computational cost.

An adaptive wavelet packet denoising algorithm for enhanced active acoustic damage detection from wind turbine blades

The development of a viable structural health monitoring (SHM) technology for the operational condition monitoring of wind turbine blades is of great interest to the wind industry. In order for any SHM technology to achieve the technical readiness and performance required for an operational implementation, advanced signal processing algorithms need to be developed to adaptively remove noise and retain the underlying signals of interest that describe the damage-related information. The wavelet packet transform decomposes a measured time domain signal into a time-frequency representation enabling the removal of noise that may overlap with the signal of interest in time and/or frequency. However, the traditional technique suffers from several assumptions limiting its applicability in an operational SHM environment, where the noise conditions commonly exhibit erratic behavior. Furthermore, an exhaustive number of options exist when selecting the parameters used in the technique with limited guidelines that can help select the most appropriate options for a given application. Appropriately defining the technique tends to be a daunting task resulting in a general avoidance of the approach in the field of SHM.This work outlines an adaptive wavelet packet denoising algorithm applicable to numerous SHM technologies including acoustics, vibrations, and acoustic emission. The algorithm incorporates a blend of non-traditional approaches for noise estimation, threshold selection, and threshold application to augment the denoising performance of real-time structural health monitoring measurements. Appropriate wavelet packet parameters are selected through a simulation considering the trade-off between signal to noise ratio improvement and amount of signal energy retained. The wavelet parameter simulation can be easily replicated to accommodate any SHM technology where the underlying signal of interest is known, as is the case in most active-based approaches including acoustic and wave-propagation techniques. The finalized adaptive wavelet packet algorithm is applied to a comprehensive dataset demonstrating an active acoustic damage detection approach on a ~46 m wind turbine blade. The quality of the measured data and the damage detection performance obtained from simple spectral filtering is compared with the proposed wavelet packet technique. It is shown that the damage detection performance is enhanced in all but one test case by as much as 60%, and the false detection rate is reduced. The approach and the subsequent results presented in this paper are expected to help enable advancement in the performance of several established SHM technologies and identifies the considered acoustics-based SHM approach as a noteworthy option for wind turbine blade structural health monitoring.

Vibration signal denoising for structural health monitoring by residual convolutional neural networks

In vibration based structural health monitoring (SHM), measurement noise inevitably exists in the vibration data, which significantly influences the usability and quality of measured vibration signals for structural identification and condition monitoring. As a result, there is a high demand for developing effective methods to reduce noise effect, especially in harsh and extreme environment. This paper proposes a vibration signal denoising approach for SHM based on a specialized Residual Convolutional Neural Networks (ResNet). Dropout, skip connection and sub-pixel shuffling techniques are used to improve the performance. The effectiveness and robustness of this developed approach are validated with acceleration data measured from Guangzhou New TV Tower. The results show that the proposed approach is effective in improving the quality of the acceleration data with varying levels of noises and different types of noises. Modal identifications based on signals contaminated with intensive noise and de-noised signals are conducted. Modal information of weakly excited modes masked by noise and closely spaced modes can be clearly and accurately identified from the de-noised signals, which could not be reliably identified with the original signal, indicating the effectiveness of using this developed approach for SHM. Besides white noise, a group of data contaminated with pink noise, which is not included in the training data, is also tested. Good results are obtained. The developed ResNet extracts high-level features from the vibration signal and learns the modal information of structures automatically, therefore it can well preserve the most important vibration characteristics in vibration signals, and can assist in distinguishing the physical modes from the spurious modes in structural modal identification.

Relation between the shear stress distribution and the resulting acoustic emission variation in concrete beams

Acoustic emission (AE) technology is extensively acknowledged as a promising structural health monitoring (SHM) approach; meanwhile, AE monitoring has long been criticised for a lack of quantitative results. This work revealed a strong correlation between the shear stress distribution and the resulting AE distribution, enabling accurate and real-time quantification of the shear stress states in concrete beams using only AE measurements. This paper started by introducing the AE phenomenon and the theoretical distribution of the shear stresses on a beam cross section, which revealed that the distributions of the stresses and the resulting emissions inherently correlate to each other. Twelve reinforced concrete beams were then tested, deploying devices including an AE system. The results showed that the AE source intensity distribution strongly correlated with that of the shear stresses in a beam segment when the beam was in both the working stage and the failure phase and that the matching degree of these two distribution curves varied from 79.20% to 94.12%. The correlation revealed in the paper will boost a wide range of significant AE-based scientific research and engineering applications, including SHM and non-destructive testing.

A compressive MUSIC spectral approach for identification of closely-spaced structural natural frequencies and post-earthquake damage detection

Motivated by practical needs to reduce data transmission payloads in wireless sensors for vibration-based monitoring of engineering structures, this paper proposes a novel approach for identifying resonant frequencies of white-noise excited structures using acceleration measurements acquired at rates significantly below the Nyquist rate. The approach adopts the deterministic co-prime sub-Nyquist sampling scheme, originally developed to facilitate telecommunication applications, to estimate the autocorrelation function of response acceleration time-histories of low-amplitude white-noise excited structures treated as realizations of a stationary stochastic process. Next, the standard multiple signal classification (MUSIC) spectral estimator is applied to the estimated autocorrelation function enabling the identification of structural natural frequencies with high resolution by simple peak picking in the frequency domain without posing any sparsity conditions to the signals. This is achieved by processing autocorrelation estimates without undertaking any (typically computationally expensive) signal reconstruction step in the time-domain, as required by various recently proposed in the literature sub-Nyquist compressive sensing-based approaches for structural health monitoring, while filtering out any broadband noise added during data acquisition. The accuracy and applicability of the proposed approach is first numerically assessed using computer-generated noise-corrupted acceleration time–history data obtained by a simulation-based framework examining white-noise excited structural systems with two closely-spaced modes of vibration carrying the same amount of energy, and a third isolated weakly excited vibrating mode. Further, damage detection potential of the developed method is numerically illustrated using a white-noise excited reinforced concrete 3-storey frame in a healthy and two damaged states caused by ground motions of increased intensity. The damage assessment relies on shifts in natural frequencies between the pre-earthquake and post-earthquake state. Overall, numerical results demonstrate that the considered approach can accurately identify structural resonances and detect structural damage associated with changes to natural frequencies as minor as 1% by sampling up to 78% below Nyquist rate for signal to noise ratio as low as 10dB. These results suggest that the adopted approach is robust and noise-immune while it can reduce data transmission requirements in acceleration wireless sensors for natural frequency identification and damage detection in engineering structures.