Nondestructive Structural Health Monitoring Research Articles

Machine Learning-Based Structural Health Monitoring Technique for Crack Detection and Localisation Using Bluetooth Strain Gauge Sensor Network

Within the domain of Structural Health Monitoring (SHM), conventional approaches generally are complicated, destructive, and time-consuming. It also necessitates an extensive array of sensors to effectively evaluate and monitor the structural integrity. In this research work, we present a novel, non-destructive SHM framework based on machine learning (ML) for the accurate detection and localisation of structural cracks. This approach leverages a minimal number of strain gauge sensors linked via Bluetooth Low Energy (BLE) communication. The framework is validated through empirical data collected from 3D carbon fibre-reinforced composites, including three distinct specimens, ranging from crack-free samples to specimens with up to ten cracks of varying lengths and depths. The methodology integrates an analytical examination of the Shewhart chart, Grubbs’ test (GT), and hierarchical clustering (HC) algorithm, tailored towards the metrics of fracture measurement and classification. Our novel ML framework allows one to replace exhausting laboratory procedures with a modern and quick mechanism for the material, with unprecedented properties that could provide potential applications in the composites industry.

Indirect prediction of graphene nanoplatelets-reinforced cementitious composites compressive strength by using machine learning approaches

Graphene nanoplatelets (GrNs) emerge as promising conductive fillers to significantly enhance the electrical conductivity and strength of cementitious composites, contributing to the development of highly efficient composites and the advancement of non-destructive structural health monitoring techniques. However, the complexities involved in these nanoscale cementitious composites are markedly intricate. Conventional regression models encounter limitations in fully understanding these intricate compositions. Thus, the current study employed four machine learning (ML) methods such as decision tree (DT), categorical boosting machine (CatBoost), adaptive neuro-fuzzy inference system (ANFIS), and light gradient boosting machine (LightGBM) to establish strong prediction models for compressive strength (CS) of graphene nanoplatelets-based materials. An extensive dataset containing 172 data points was gathered from published literature for model development. The majority portion (70%) of the database was utilized for training the model while 30% was used for validating the model efficacy on unseen data. Different metrics were employed to assess the performance of the established ML models. In addition, SHapley Additve explanation (SHAP) for model interpretability. The DT, CatBoost, LightGBM, and ANFIS models exhibited excellent prediction efficacy with R-values of 0.8708, 0.9999, 0.9043, and 0.8662, respectively. While all the suggested models demonstrated acceptable accuracy in predicting compressive strength, the CatBoost model exhibited exceptional prediction efficiency. Furthermore, the SHAP analysis provided that the thickness of GrN plays a pivotal role in GrNCC, significantly influencing CS and consequently exhibiting the highest SHAP value of + 9.39. The diameter of GrN, curing age, and w/c ratio are also prominent features in estimating the strength of graphene nanoplatelets-based cementitious materials. This research underscores the efficacy of ML methods in accurately forecasting the characteristics of concrete reinforced with graphene nanoplatelets, providing a swift and economical substitute for laborious experimental procedures. It is suggested that to improve the generalization of the study, more inputs with increased datasets should be considered in future studies.

Acoustic emission onset time detection for structural monitoring with U-Net neural network architecture

Acoustic Emission (AE) is a non-destructive structural health monitoring technique, which studies elastic waves emitted during crack formation. Utilizing piezoelectric sensors, these waves are converted into electrical signals for subsequent analysis, offering insights into crack propagation and structural durability. This study focuses on the identification of AE signal onset times, crucial for determining crack locations. Conventional methods often encounter challenges with background noise, prompting the need for innovative approaches. Leveraging a U-Net neural network, specialized in segmentation tasks, onset time identification is approached as a one-dimensional segmentation challenge. Through training and testing on Pencil Lead Break (PLB) test data, commonly used in AE evaluations, the effectiveness of the method is demonstrated even with continuous signals, suggesting potential applicability in real-time monitoring.

Engineering Reliability-Based Condition Assessment for Stay Cables Using Non-Destructive Interferometric Radar

Structural health monitoring (SHM) of cable-stayed bridges requires periodic assessment for the deteriorating stay cables to ensure a long-term service life of the bridge. However, conducting the non-destructive SHM for operating cable-stayed bridges and analyzing the safety statuses of all the working cables are still challenging, due to the lack of in situ cable data for previously constructed bridges. This study developed an innovative framework of health condition assessment for stay cables based on cable vibration frequencies from an interferometric radar (IBIS-FS) using engineering reliability analysis (ERA). Taking a cable-stayed bridge in Victoria, Australia as a target structure for the case study, it shows that the presented framework can remotely monitor the accurate real-time load bearing conditions of stay cables by calculating tension forces, and effectively assess their health conditions. The results show that the natural frequency (up to the fifth mode) of a healthy cable remains constant under different external loadings but varies for damaged cables. The measured reliability index of all the stay cables is higher than the safety threshold factor at ultimate limit states, while one carries tension force higher than the maximum design load (lower than the minimum breaking load) and other three cables need to be monitored regularly due to their low reliability indices. This is attributed to an integrated effect of applied tension force, cable diameters, and minimum breaking loads.

Observation of an exceptional point with an LR-shunted resonator

Non-Hermitian physical systems have enriched wave control abilities and dynamics of wave matter, especially around exceptional points (EPs). In this work, we theoretically, numerically, and experimentally studied an LR-shunted mechanical resonator for the observation of the exceptional point. The shunted resonator was composed of a mechanical resonator with a bonded piezoelectric patch and an LR-shunted circuit. A theoretical model for the LR-shunted resonator was first developed, and the exceptional point was identified from this theoretical model via a parameter space study. It was found that by varying the shunting circuit parameters, i.e., inductance and resistance, such a simple design supports the non-Hermitian degeneracy, namely the exceptional point. At the vicinity of the EP, the LR-shunted mechanical resonator had a square-root dependence on the external perturbation related to the nominal resistance. Next, we designed and fabricated a shunted resonator to verify our design through numerical analysis and experimental measurements. Moreover, our proposed LR-shunted resonator provides the possibility to dynamically encircle an exceptional point due to external stimuli. Our approach sheds light on the designs of chiral effect systems and dynamically tailoring losses in elasticity, and it enables alternative solutions for nondestructive structural health monitoring with enhanced sensitivity.

Self-learning locally-optimal hypertuning using maximum entropy, and comparison of machine learning approaches for estimating fatigue life in composite materials of the aerospace industry

Applications of Structural Health Monitoring (SHM) combined with Machine Learning (ML) techniques enhance real-time performance tracking and increase structural integrity awareness of civil, aerospace and automotive infrastructures. This SHM-ML synergy has gained popularity in the last years thanks to the anticipation of maintenance provided by arising ML algorithms and their ability of handling large quantities of data and considering their influence in the problem.In this paper we develop a novel ML nearest-neighbors-alike algorithm based on the principle of maximum entropy to predict fatigue damage (Palmgren–Miner index) in composite materials by processing the signals of Lamb Waves – a non-destructive SHM technique – with other meaningful features such as layup parameters and stiffness matrices calculated from the Classical Laminate Theory (CLT). The full data analysis cycle is applied to a dataset of delamination experiments in composites. The predictions achieve a good level of accuracy, similar to other ML algorithms, e.g. Neural Networks or Gradient-Boosted Trees, and computation times are of the same order of magnitude.The key advantages of our proposal are: (1) The automatic determination of all the parameters involved in the prediction, so no hyperparameters have to be set beforehand, which saves time devoted to hypertuning the model and also represents an advantage for autonomous, self-supervised SHM. (2) No training is required, which, in an online learning context where streams of data are fed continuously to the model, avoids repeated training—essential for reliable real-time, continuous monitoring.

Real-time monitoring of residual strength in corroding steel reinforcement using ultrasonic-guided waves and multi-physics modelling

Corrosion-induced material strength degradation is a major threat to the service life of reinforced concrete (RC) structures. Continued mass loss and pit formation after the initiation of corrosion in reinforcing steel consequently results in strength degradation. In this study, a non-destructive structural health monitoring (SHM) approach using ultrasonic guided waves with multi-physics modelling to monitor and forecast strength degradation of corroding steel is presented. Surface-bonded piezoelectric wafer transducers (PWTs) are used for actuating and sensing of guided waves in order to correlate the group velocity and amplitude of flexural wave modes with the strength reduction. The major findings of the work are presented in the form of empirical relationships between guided wave characteristics with reduced yield, ultimate, and buckling strength of corroded steel. It is observed that for a rebar of 12 mm diameter having slenderness ratio between 8 and 10, a 10% increase in amplitude and 8.44% increase in group velocity of flexural mode for corroding steel indicates a corresponding reduction of 10.278% in yield strength, 10.277% in ultimate strength, and 4.45% in buckling strength in progression phase of corrosion. A multi-physics numerical model of corrosion in RC structures that considers the effect of local exposure and environment is then elucidated with deviation less than 6.34% from experimental results in order to forecast the mass loss. The presented data fusion strategy, which combines guided wave-based monitoring and multi-physics modelling finds valuable application in assessing the current state of degradation in structural performance and strength-based service life prediction.

Multiple Damage Detection in PZT Sensor Using Dual Point Contact Method

Lead Zirconate Titanate (PZT) is used to make ultrasound transducers, sensors, and actuators due to its large piezoelectric coefficient. Several micro-defects develop in the PZT sensor due to delamination, corrosion, huge temperature fluctuation, etc., causing a decline in its performance. It is thus necessary to identify, locate, and quantify the defects. Non-Destructive Structural Health Monitoring (SHM) is the most optimal and economical evaluation method. Traditional ultrasound SHM techniques have a huge impedance mismatch between air and solid material, and most of the popular signal processing methods define time series signals in only one domain, which provides sub-optimal results for non-stationary signals. Thus, to improve the accuracy of detection, the point contact excitation and detection method is implemented to determine the interaction of ultrasonic waves with micro-scale defects in the PZT. The signal generated from this method being non-stationary in nature, it requires signal processing with changeable resolutions at different times and frequencies. The Haar Discrete Wavelet Transformation (DWT) is applied to the time series data obtained from the coulomb coupling setup. Using the above process, defects up to 100 μm in diameter could be successfully distinguished.

Performance assessment of steel truss railway bridge with curved track

Non-destructive Structural Health Monitoring techniques can be incorporated into bridge integrity management by assessing structural conditions. This paper describes a performance assessment of a steel truss railway bridge in Bratislava using vibration-based techniques as a further part of maintenance in addition to standard visual inspections. To obtain the necessary data, a multipurpose measuring system was used. Various types of data were measured, e.g. accelerations, strains, and displacements. The advantage of the multipurpose measuring system was that the traffic over the bridge was not restricted, even though the bridge carries only a single curved track. Two test campaigns were conducted to assess the performance of the bridge. One campaign was devoted to measuring ambient vibrations in order to perform the operational modal analysis, and the second was carried out to measure strains and displacements during a train passage. The results show a successful system identification of the structure using ambient vibrations; and a finite element model was verified and validated by a comparison of strains and displacements, as well as by modal parameters. According to the results obtained, the structural health of the investigated bridge was satisfactory.

An artificial intelligence-based conductivity prediction and feature analysis of carbon fiber reinforced cementitious composite for non-destructive structural health monitoring

Carbon fiber reinforced cementitious composite (CFRCC) possesses stable and controllable electrical properties for non-destructive structural health monitoring (SHM) by electrical resistivity measurement (ERM) to determine the durability of structures. It is necessary to build an accurate and robust prediction model of CFRCC’s resistivity detected by ERM under the influence of complex factors to achieve better SHM performance. To address this problem, the present study developed an integrated modelling approach with multiple machine learning and optimization algorithms to identify the best model for predicting CFRCC’s resistivity. A dataset containing 602 experimental instances was constructed based on information accessed in the research literature. Results show that the XGBoost model tuned by Bayesian optimization had the best predictive performance with the largest R2 scores (0.95 and 0.96) and was thus proposed as the prediction model. Weight scores of input factors were given by the proposed model revealing that CFRCC’s resistivity highly depended on carbon fiber and sand content in the composite. The influences of critical factors on model output were further quantified by Shapley Additive Explanations (SHAP) algorithm. The developed method can be applied to aid the design optimization for CFRCC and the non-destructive SHM of CFRCC-based structures by ERM.

Identification of fracture damage characteristics in ultra-high performance cement-based composite using digital image correlation and acoustic emission techniques

The study of fracture mechanism and damage of ultra-high performance fiber reinforced cement-based composite (UHPFRCC) is of great significance for material performance evaluation and nondestructive structural health monitoring. In this paper, three-point bending fracture tests of UHPFRCC with fiber content of 1% and 2% were carried out, and the digital image correlation (DIC) and acoustic emission (AE) technique were used to monitor the fracture simultaneously. The microcrack area of UHPFRCC1 (UHPFRCC with 1% fiber content) decreases gradually with the loading process, and the microcrack area of UHPFRCC2 (UHPFRCC with 2% fiber content) increases first and then decreases with the loading process. The macrocrack area of UHPFRCC1 and UHPFRCC2 increases gradually with the same growth trend. The matrix failure and fiber/matrix debonding behavior of UHPFRCC were identified effectively based on the unsupervised k-means clustering method, and the fiber/matrix debonding behavior was obvious when the AE amplitude was greater than 60 dB. Meanwhile, the clustering results are also verified by the evolution of FPZ length. Finally, an efficient identification model of UHPFRCC matrix failure and fiber/matrix debonding behavior based on supervised KNN algorithm is proposed, which provides theoretical support for nondestructive structural health monitoring in practical engineering.

The effect of external load on ultrasonic wave attenuation in steel bars under bending stresses

The stress state in deformed solids has a significant impact on the attenuation of an ultrasonic wave propagating through the medium. Measuring a signal with certain attenuation characteristics can therefore provide useful diagnostic information about the stress state in the structure. In this work, basic principles behind a novel attenuation-based diagnostic framework are introduced. An experimental study on steel bars under three-point bending was carried out, and finite element analyses were used to numerically model the experiments. Obtained test results showed a strong correlation between the external load and the ultrasonic signal energy, which decreases with increasing load. A similar but positive correlation appeared between the level of attenuation of longitudinal ultrasonic wave signals and the external load, which allowed for efficient estimation of the mid-span bending moment. Upon proper calibration of testing equipment, the change in ultrasonic signal energy can therefore be used as an indicator of the external load level. As a result, this effect has potential applications in non-destructive structural health monitoring frameworks.

Multi-objective design optimization for graphite-based nanomaterials reinforced cementitious composites: A data-driven method with machine learning and NSGA-Ⅱ

Graphite-based nanomaterials (GNs) are promising conductive fillers for producing highly effective electrically conductive cementitious composites (ECCC) and promoting non-destructive structural health monitoring (SHM) methods. Since acceptable mechanical strength and electrical resistivity are both required, the design of GN-reinforced cementitious composites (GNRCC) is a complicated multi-objective optimization problem (MOOP). The present study proposes a comprehensive data-driven method to address this multi-objective design optimization (MODO) issue for GNRCC using machine learning (ML) techniques and non-dominated sorting genetic algorithm (NSGA-Ⅱ). First, prediction models of uniaxial compressive strength (UCS) and electrical resistivity (ER) of GNRCC are established by Bayesian-tuned XGBoost with prepared experimental datasets. The results show that they have excellent performance in predicting both properties with high R2 (0.95 and 0.92, 0.99 and 0.98) and low mean absolute error (MAE) scores (1.24 and 3.44, 0.15 and 0.22). The influence of critical features on GNRCC’s properties are quantified by ML theories. This helps determine the variables to be optimized and define their constraints for the MODO. Finally, the MODO program is developed on the basis of NSGA-Ⅱ. It optimizes GNRCC’s properties of UCS and ER simultaneously with the proposed prediction models as objective functions. It successfully achieves a set of Pareto solutions, which can facilitate appropriate parameters selections for the GNRCC design.

Metal/Carbon-Fiber Hybrid Composites—Damage Evolution and Monitoring of Isothermal Fatigue at Low and Elevated Temperatures

Carbon-fiber-reinforced polymers (CFRPs) are the standard lightweight composite material for structural applications in aviation. The addition of metallic fibers to CFRPs to form metal/carbon-fiber hybrid composites (MCFRPs) has been shown to improve the elastic and plastic properties and to enable a non-destructive method for structural health monitoring over the material’s service life. In this paper, the results from the fatigue experiments on these hybrid composites at −55, 25 and 120 °C are discussed. Multidirectional CFRP and MCFRP laminates, fabricated using the autoclave method, were tested and compared under different fatigue loading conditions, while being simultaneously monitored for temperature and electrical resistance. Magnetic phase measurements were additionally carried out for the chosen metastable austenitic steel fibers in the MCFRPs. The results show that the improved ductility of the hybrid composite due to the presence of the steel fibers leads to better performance under fatigue loads and a less-brittle failure behavior. Based on the chemical composition of the metastable austenitic steel fibers, a temperature and plastic deformation-dependent phase transformation was observed, which could potentially lead to a method for non-destructive structural health monitoring of the hybrid composite over its service life.

A novel damage index for reinforced concrete beam-column joint using acoustic emission technique

Acoustic Emission (AE) is a non-destructive Structural Health Monitoring (SHM) technique used to detect the presence and movement of defects in reinforced concrete structures. Various AE based techniques help to identify the occurrence of damage, but there is a need to develop a mechanism for the evaluation of damage index based on AE parameters. This study investigates the ability of AE energy parameters to detect damage to a reinforced concrete (RC) beam-column joint (BCJ). An experimental study has been conducted to test BCJ under cyclic load, and damage progression has been monitored using AE parameters. The results showed a significant increase in energy dissipated and AE energy when a BCJ is reinforced with ductile detailing. Further, a relationship has been developed between energy dissipated and AE energy to determine the damage index. A novel damage index based on AE energy has been developed for reinforced concrete beam-column joints.

Guided Wave Studies for Enhanced Acoustic Emission Inspection

ABSTRACT Fundamental elements of wave mechanics are covered with respect to Acoustic Emission (AE) analysis in Nondestructive Evaluation (NDE) and Structural Health monitoring (SHM). Emphasis is placed on aspects of Ultrasonic Guided Waves that travel in a structure due to elastic wave emissions from a defect source as defects continue to grow. The AE Method is based on the emission of elastic waves from a particular source as failure is initiated – crack, corrosion, leak, and so on. The AE signal received is a function of the source orientation and location as well as wave velocities in the structure, sensor types and positions, arrival time differences, thicknesses of the structure, and specimen structural variations. Such topics as Guided wave physical and theoretical considerations, Ultrasonic Guided wave accomplishments, and enhanced AE by considering various guided wave concepts are discussed. The topic of Acousto-Ultrasonics and its impact on guided wave understanding is also reviewed. This paper illustrates how principles in Ultrasonic Guided waves can be applied to Acoustic Emission. Besides basic issues, Acoustic Emission enhancement possibilities are based on recent studies of Ultrasonic Guided Waves and the use of Shear Horizontal guided waves in Acoustic Emission along with an omni-directional shear horizontal wave transducer. Editor’s note: Joseph L. Rose, PhD, is the recipient of the 2021 ASNT Research Recognition for Sustained Excellence. Rose presented on “Guided Wave Studies for Enhanced Acoustic Emission Inspection” during the ASNT Research Symposium which was held 27–29 April 2021.

Energy Harvesting Technologies for Structural Health Monitoring of Airplane Components-A Review.

With the aim of increasing the efficiency of maintenance and fuel usage in airplanes, structural health monitoring (SHM) of critical composite structures is increasingly expected and required. The optimized usage of this concept is subject of intensive work in the framework of the EU COST Action CA18203 “Optimising Design for Inspection” (ODIN). In this context, a thorough review of a broad range of energy harvesting (EH) technologies to be potentially used as power sources for the acoustic emission and guided wave propagation sensors of the considered SHM systems, as well as for the respective data elaboration and wireless communication modules, is provided in this work. EH devices based on the usage of kinetic energy, thermal gradients, solar radiation, airflow, and other viable energy sources, proposed so far in the literature, are thus described with a critical review of the respective specific power levels, of their potential placement on airplanes, as well as the consequently necessary power management architectures. The guidelines provided for the selection of the most appropriate EH and power management technologies create the preconditions to develop a new class of autonomous sensor nodes for the in-process, non-destructive SHM of airplane components.

Deep‐learning‐based guided wave detection for liquid‐level state in porcelain bushing type terminal

The structural security of civil energy equipment is significant for the steady operation of power supply system, and porcelain bushing type terminal is a typical energy equipment that needs long-term monitoring. As a nondestructive structural health monitoring method, ultrasonic guided wave (UGW) technology is extremely suitable for state detection of energy equipment. However, most current UGW methods still need to manually select the guided wave features, which rely heavily on the guidance of expert experience. This article presents a deep-learning method to directly utilize original-guided wave signals to quantitatively detect the liquid-level state. Firstly, the original signals were fed into convolutional autoencoder (CAE) to catch the low-dimension representation and realize the automatic feature extraction. Then, the low-dimension representations were orderly input into the long short-term memory (LSTM) recurrent neural network for liquid-level regression. In feature extraction step, CAE method can effectively extract the useful features and remove the interference and signal distortion. In regression step, both the accuracy and the robustness of proposed method are better than backpropagation network and convolutional neural network. Experimental results show that proposed CAE-LSTM method can accurately inspect the liquid level by original signals and implement maintenance monitoring.

Multipoint and Energy-Equal Fiber-Optic Laser-Ultrasonic Actuator Based on Peanut-Shaped Structures

Laser-ultrasonic excitation holds promising prospects in nondestructive testing and structural health monitoring. This paper proposes a multipoint, energy-equal laser-ultrasonic actuator based on multiple peanut-shape (PNS) structures in a single mode fiber link. The PNS structures were fabricated by fiber micro-matching operations using a fiber splicer. The PNSs effectively couple the laser energy from the core mode into the cladding modes in the fiber. Moreover, the coupling ratio can be adjusted by controlling the taper diameter of the PNSs. For energy control, multiple PNS structures are fabricated and connected in order from small to large coupling ratios. Experiments were performed on an actuator with four PNSs of different taper diameters (138.71 $\mu \text{m}$ , 143.42 $\mu \text{m}$ , 158.23 $\mu \text{m}$ , and 180.19 $\mu \text{m}$ ) distributed along a single fiber. With its simple structure, energy control, and multipoint functionality, this all-fiber laser-ultrasonic actuator shows great potential in future multipoint ultrasound applications.

The Quality Assessment of Different Geolocalisation Methods for a Sensor System to Monitor Structural Health of Monumental Objects.

Cultural heritage objects are affected by a wide range of factors causing their deterioration and decay over time such as ground deformations, changes in hydrographic conditions, vibrations or excess of moisture, which can cause scratches and cracks formation in the case of historic buildings. The electromagnetic spectroscopy has been widely used for non-destructive structural health monitoring of concrete structures. However, the limitation of this technology is a lack of geolocalisation in the space for multispectral architectural documentation. The aim of this study is to examine different geolocalisation methods in order to determine the position of the sensor system, which will then allow to georeference the results of measurements performed by this device and apply corrections to the sensor response, which is a crucial element required for further data processing related to the object structure and its features. The classical surveying, terrestrial laser scanning (TLS), and Structure-from-Motion (SfM) photogrammetry methods were used in this investigation at three test sites. The methods were reviewed and investigated. The results indicated that TLS technique should be applied for simple structures and plain textures, while the SfM technique should be used for marble-based and other translucent or semi-translucent structures in order to achieve the highest accuracy for geolocalisation of the proposed sensor system.