Structural Health Monitoring Techniques Research Articles

Machine Learning Techniques for Structural Health Monitoring of Concrete Structures: A Systematic Review

Machine Learning Techniques for Structural Health Monitoring of Concrete Structures: A Systematic Review

On the modeling of circular piezoelectric transducers for wave propagation-based structural health monitoring applications

Structural Health Monitoring (SHM) techniques have an important role in the performance of mechanical structures. In particular, piezoelectric transducers (PZT) have been employed for establishing damage detection processes in SHM systems. However, despite their wide use, there is limited information in the literature regarding important characteristics for SHM applications, such as the shape, dimensions and frequencies to use, which can affect the performance of the system used to detect the damage. This article investigates longitudinal and flexural waves for applications using circular piezoelectric transducers bonded to thin plates. The methodology allows one to obtain optimal frequencies to create and capture both types of waves and understand the influence of the geometric characteristics of these transducers on the damage detection. Numerical simulations are carried out to show that a change in the parameters of the piezoelectric transducers can maximize the waves amplitudes incoming at the sensor. Results from experimental tests are presented to demonstrate the proposed methodology. New equations are introduced and they compute output voltage and determine optimal frequencies to monitor the structure. The findings contribute to establishing a more efficient design of a damage detection process involving plate-like structures in SHM systems based on wave propagation.

An ultrasonic guided waves based prognostic approach for predictive maintenance: Experimental study cases

This paper concerns the use of ultrasonic guided waves (UGW) in predictive maintenance (PM). Generally, PM is based on machine or process parameters such as current, voltage, temperature and pressure. However, research studies on the use of structural health monitoring (SHM) techniques in PM are seldom (to the best knowledge of the authors). Hence, the main purpose of the present study is to investigate the feasibility of using UGW, in order to reinforce the PM of a given installation. An approach that combines diagnostic and prognostic models is developed. The predictive model is hybrid and based on Empirical Mode Decomposition (EMD) and Autoregressive Integrated Moving Average (ARIMA). The built model is applied on data collected from two kinds of structures, namely flexible and rigid risers. Residual useful life (RUL) of each structure is estimated and compared with its simulated actual value. The paper provides a fruitful discussion about the applicability of UGW for prognostic as this is an important step in SHM and there is still a lack of research related to this topic.

BIM oriented applications of structural health monitoring based on magnified digital image correlation point-clouds

Three-dimensional Digital Image Correlation (3D-DIC) gained popularity as a line-of-sight, high-resolution Structural Health Monitoring (SHM) technique. Nonetheless, 3D-DIC is still impractical when the natural frequencies and the mode shapes of the structure have to be correlated. Recently, the integration of Motion Magnification (MM) as a pre-processing tool for 3D-DIC has been explored to enhance the capabilities of DIC for structural analysis and defined a Magnified DIC (M-DIC) workflow. However, such applications do not take full advantage of the potential of the 3D-DIC to produce point-clouds informed of the dynamic properties of a structure. Hence, the definition of a straightforward workflow to manage 3D-DIC point-clouds for real-world applications should be addressed. Moreover, the large amount of raw output coming from MM and 3D-DIC combination requires appropriate tools to systematically visualise and analyse the data. In this context, Building Information Modelling (BIM) has emerged as a robust repository as well as a data management tool to store and exchange data related to the built environment, including long sequences of SHM data. This research proposes a novel Cloud to Model (C2M) tool called M-DIC2BIM that translates the Scalar Fields (SFs) of the M-DIC point-cloud directly into the faces of the BIM model instances. As a case study, a three-storey aluminium frame structure excited with broadband noise is used to validate the developed methodology. The comparison among undamaged and damaged configurations tests made the evolution of the frame's Operational Deflection Shapes (ODSs) computable by quantifying structural changes in its global and local behaviour and mapping them onto the BIM instances.

Cracking and Fiber Debonding Identification of Concrete Deep Beams Reinforced with C-FRP Ropes against Shear Using a Real-Time Monitoring System.

Traditional methods for estimating structural deterioration are generally costly and inefficient. Recent studies have demonstrated that implementing a network of piezoelectric transducers mounted to critical regions of concrete structural members substantially increases the efficacy of the structural health monitoring (SHM) method. This study uses a recently developed electro-mechanical-admittance (EMA)-based SHM system for real-time damage diagnosis of carbon FRP (C-FRP) ropes installed as shear composite reinforcement in RC deep beams. The applied SHM technique uses the frequency response measurements of a network of piezoelectric lead zirconate titanate (PZT) patches. The proposed strengthening methods using C-FRP ropes as ETS and NSM shear reinforcement and the applied anchorage techniques significantly enhanced the strength and the overall performance of the examined beams. The retrofitted beams exhibited increased shear capacity and improved post-peak response with substantial ductility compared with the brittle failure of the non-strengthened specimens. The health condition and the potential debonding failure of the applied composite fiber material were also examined and quantified using the proposed SHM technique. Damage quantification of C-FRP ropes is achieved by comparing and assessing the values of several statistical damage indices. The experimental results demonstrated that the proposed monitoring system successfully diagnosed the region where the damage occurred by providing early warning of the forthcoming critical shear cracking of concrete and C-FRP rope debonding failures. Furthermore, the internal PZT transducers showed sound indications of the C-FRP rope's health condition, demonstrating a direct correlation with the mechanical performance of the fibers.

Combining Structural Health Monitoring and Earthquake Early Warning Techniques for an Enhanced Monitoring System

Advancements in Civil Engineering & Technology Combining Structural Health Monitoring and Earthquake Early Warning Techniques for an Enhanced Monitoring System Shieh-Kung Huang* Assistant Professor, Department of Civil Engineering, National Chung-Hsing University, Taiwan *Corresponding author:Shieh-Kung Huang, Assistant Professor, Department of Civil Engineering, National Chung-Hsing University, Taiwan Submission: December 16, 2022;Published: January 05, 2023 DOI: 10.31031/ACET.2023.05.000613 ISSN: 2639-0574 Volume5 Issue3

A Review of Data Management and Visualization Techniques for Structural Health Monitoring Using BIM and Virtual or Augmented Reality

A Review of Data Management and Visualization Techniques for Structural Health Monitoring Using BIM and Virtual or Augmented Reality

Application of Multi-Temporal InSAR (MT-InSAR) for structural monitoring: the case study of Scrovegni Chapel in Padova

In the last years, the use of Multi-Temporal Interferometric Synthetic Aperture Radar (MT-InSAR) technique for structural health monitoring (SHM) of buildings and infrastructures has spread. Compared to traditional monitoring methods, MT-InSAR technique involves a number of advantages, e.g. perform back-analysis and extend the monitoring to the surrounding area, that makes it a very cost-effective tool. On the other hand, some drawbacks still have to be fully overcome, such as the issues related to the geocoding of the measurement points and the noisiness in the time series. In this article, the application of MT-InSAR monitoring to a case study is presented. The structure of interest is the Scrovegni Chapel in Padova, recently declared a world heritage site by UNESCO as it contains numerous frescos executed by Giotto. Using images acquired by COSMO-SkyMed constellation, the deformation of the area surrounding the Chapel was estimated, by means of spatial interpolation algorithms. In addition, time series referred to measurement points located on the rooftop were analyzed and compared to the environmental data (temperature and humidity) acquired by on-site physical sensors already installed.

Structural Health Monitoring and Dynamic Identification of the Historical Town-Hall of Borgo Val di Taro

A good structural knowledge of heritage buildings is a key factor for their conservation and for the planning of retrofit interventions, especially in high seismic risk areas, like Italy. Within this context, structural health monitoring techniques, like those based on ambient vibrations under operating conditions, represent powerful tools for evaluating the dynamic behavior of existing structures. The presented methodology is particularly interesting for historic and monumental buildings due to its non-invasiveness. This work illustrates the installation of a monitoring system in the historical masonry Town-Hall of Borgo Val di Taro, dating back to the fifteen century, for its ambient vibration modal identification. The so collected data are then used for the construction of a refined numerical model for the simulation of the dynamic response of the building, and for the updating of its parameters. A comparison between numerical and experimental results is finally provided.

Structural health monitoring of Cerasol® roof tiles with embedded FBG sensors

ABSTRACT There are currently little widely used structural health monitoring (SHM) techniques for structural adhesive joints. In this study, integrated fibre Bragg grating (FBG) sensors are used to examine the structural integrity and reliability of functional adhesive joints. There is an even greater demand for structural monitoring during operation or quality control of adhesive joints during production as more industries embrace high performance structural adhesives in fail-safe designs. Smartroof® photovoltaic (PV) roof tiles are used in this study to validate a real-time health monitoring of the adhesive connection between the PV glass plate and the polymer frame. A structural epoxy and a modified siloxane (MS) polymer are assessed. To assure measurement accuracy and to provide a better understanding of the strain sensing mechanism, the data is qualitatively compared with a finite element model. It is demonstrated that using embedded FBG sensors is accurate and adequate for in situ real-time structural health monitoring. Nevertheless, it remains difficult to ensure and manage the fibre’s position in the adhesive layer.

Stress monitoring capability of magnetostrictive Fe–Co fiber/glass fiber reinforced polymer composites under four-point bending

Many structural health monitoring (SHM) techniques have been investigated for damage detection in woven glass fiber reinforced polymer (GFRP) laminates. Recently, the GFRP composites integrated with sensors have received attention because the composite material can transmit information about the structural condition during operation. Magnetostrictive materials are considered as feasible candidates to realize the contactless SHM techniques by exploiting the Villari effect, but the theoretical modeling to correlate a magnetostrictive response with structural conditions is a critical issue. In this study, the analytical procedure considering the mechanics of materials and electromagnetism was proposed to model the magnetic induction by the Villari effect of magnetostrictive GFRP laminates under bending. The magnetostrictive Fe–Co fiber/GFRP composites were then developed, and the four-point bending tests were carried out to evaluate the fabricated composites’ stress monitoring capability. The magnetic flux density behavior corresponded to the bending stress fluctuation. The maximum magnetic flux density change was 70.7 mT subjected to the peak bending stress of 158 MPa. The analytical solutions showed reasonable agreement with the experimental results. The applied stress and measured magnetic flux density were correlated by the theoretical models. Thus, these results suggest an important step in realizing the novel contactless SHM technique utilizing magnetostrictive materials.

Structural integrity assessment of a full-scale adhesively-bonded bi-material joint for maritime applications

The present study proposes a comprehensive integrity assessment approach for a full-scale adhesively-bonded bi-material joint for maritime applications. The joint represents a cross-section of the bond-line connection of a ship with a steel hull and a sandwich composite superstructure. The full-scale joint consists of a sandwich composite core adhesively bonded to two U-shaped steel brackets. The joint was subjected to a quasi-static loading profile including 6 load cycles up to the final failure. Each load cycle was followed by a dwell segment holding the joint at the maximum displacement for 30 s and then unloading to 50% of the maximum displacement. Three Structural Health Monitoring (SHM) techniques including Acoustic Emission (AE), Fiber Optic Sensor (FOS), and Digital Image Correlation (DIC) were employed during the test to assess the damage state of the joint. Moreover, a Finite Element Model (FEM) was developed to simulate the evolution behavior of different damage mechanisms in the joint and the FE results were compared against the experimental findings. The obtained results showed that the integration of all the employed techniques could successfully detect the damage initiation, assess the severity of the damage, localize the critical regions of the joint, and distinguish the different damage mechanisms.

A review of bridge health monitoring based on machine learning

This paper reviews structural health monitoring (SHM) techniques of bridge structures based on machine learning (ML) algorithms. Regular inspections and the use of non-destructive testing are still the common damage-detection methods; however, they are susceptible to subjectivity and human error and involve prolonged duration. With emerging technologies such as artificial intelligence and the development of wireless sensors, SHM has shifted from offline model-driven damage detection to online/real-time data-driven damage detection. In this paper, both supervised and unsupervised ML algorithms are examined to determine which of the latest methods would be the most suitable and effective for the SHM of bridge structures. This review paper investigates recent studies on data acquisition, data imputation, data compression, feature extraction and pattern recognition using supervised/unsupervised ML algorithms.

Hybrid Fiber Optic Cable for Strain Profiling and Crack Growth Measurement in Rock, Cement, and Brittle Installation Media.

Brillouin scattering-based distributed fiber optic sensing (DFOS) technologies such as Brillouin optical time domain reflectometry (BOTDR) and Brillouin optical time domain analysis (BOTDA) have broad applicability for the long term and real-time monitoring of large concrete structures, underground mine excavations, pit slopes, and deep subsurface wellbores. When installed in brittle media, however, the meter scale spatial resolution of the BOTDR/A technology prohibits the detection or measurement of highly localized deformations, such as those which form at or along cracks, faults, and other discontinuities. This work presents a novel hybrid fiber optic cable with the ability to self-anchor to any brittle installation media without the need for manual installation along fixed interval points. Laboratory scale testing demonstrates the ability of the hybrid fiber optic cable to measure strains across highly localized deformation zones in both tension and shear. In addition, results show the applicability of the developed technology for strain monitoring in high displacement environments. Linear relationships are proposed for use in estimating the displacement magnitude along discontinuities in brittle media from strain signals collected from the hybrid fiber optic cable. The hybrid fiber optic cable has broad potential applications, such as geomechanical monitoring in underground mines, surface pits, large civil infrastructure projects, and deep subsurface wellbores. The benefits of fiber optic sensing, such as the intrinsic safety of the sensors, the long sensing range, and real time capabilities make this a compelling technique for long term structural health monitoring (SHM) in a wide range of industrial and civil applications.

Example of using long-term monitoring and non-invasive measurements for Structural Health Monitoring. Cause study of St. Anna's Church in Warsaw

Heritage objects are affected by many factors that affect degradation over time and damage to historical sites. For this reason, it is necessary to carry out continuous monitoring using non-invasive techniques for Structural Health Monitoring and plan the conservation works. This article aims to present the results of a project that involved long-term monitoring using surveying techniques such as precise tacheometric measurements, precise levelling, inclinometric measurements and crack measurements using crack gauges. This paper presents the results of multisensor monitoring of displacements and cracks in the building, which contributed to the conservation efforts and prevent works. A research object is St. Anne's Church in Warsaw. It is one of the few in Warsaw that wasn't destroyed during World War II.

Experimental and Computational Vibration Analysis for Diagnosing the Defects in High Performance Composite Structures Using Machine Learning Approach

Delamination in laminated structures is a concern in high-performance structural applications, which challenges the latest non-destructive testing techniques. This study assesses the delamination damage in the glass fiber-reinforced laminated composite structures using structural health monitoring techniques. Glass fiber-reinforced rectangular laminate composite plates with and without delamination were considered to obtain the forced vibration response using an in-house developed finite element model. The damage was diagnosed in the laminated composite using machine learning algorithms through statistical information extracted from the forced vibration response. Using an attribute evaluator, the features that made the greatest contribution were identified from the extracted features. The selected features were further classified using machine learning algorithms, such as decision tree, random forest, naive Bayes, and Bayes net algorithms, to diagnose the damage in the laminated structure. The decision tree method was found to be a computationally effective model in diagnosing the delamination of the composite structure. The effectiveness of the finite element model was further validated with the experimental results, obtained from modal analysis using fabricated laminated and delaminated composite plates. Our proposed model showed 98.5% accuracy in diagnosing the damage in the fabricated composite structure. Hence, this research work motivates the development of online prognostic and health monitoring modules for detecting early damage to prevent catastrophic failures of structures.

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.

Deep neural network-based structural health monitoring technique for real-time crack detection and localization using strain gauge sensors

Structural health monitoring (SHM) techniques often require a large number of sensors to evaluate and monitor the structural health. In this paper, we propose a deep neural network (DNN)-based SHM method for accurate crack detection and localization in real time using a small number of strain gauge sensors and confirm its feasibility based on experimental data. The proposed method combines a DNN model with principal component analysis (PCA) to predict the strain field based on the local strains measured by strain gauge sensors located rather sparsely. We demonstrate the potential of the proposed technique via a cyclic 4-point bending test performed on a composite material specimen without cracks and seven specimens with different lengths of cracks. A dataset containing local strains measured with 12 strain gauge sensors and strain field measured with a digital image correlation (DIC) device was prepared. The strain field dataset from DIC is converted to a smaller dimension latent space with a few eigen basis via PCA, and a DNN model is trained to predict principal component values of each image with 12 strain gauge sensor measurements as input. The proposed method turns out to accurately predict the strain field for all specimens considered in the study.

Acoustic Emission Source Characterisation during Fatigue Crack Growth in Al 2024-T3 Specimens.

While acoustic emission (AE) testing can be used as a valuable technique in structural health monitoring and non-destructive testing, little research has been conducted to establish its sources, particularly in 2024-T3 aluminium alloys. The major contribution of this work is that it provides a method to obtain a better linear relationship of count rate with crack growth rate based on waveform. This paper aims to characterise AE sources by synchronising the AE waveforms with load levels and then to propose possible dominant frequency ranges. The AE waveforms during fatigue crack growth in edge-notched 2024-T3 aluminium specimens, from an initial crack length of 10 mm to 70 mm, were collected at two different load ratios R = 0.125 and 0.5. At the same time, the crack growth rate was determined using thermal imaging and associated control software. The AE waveforms obtained were processed using the fast Fourier transform. It was shown that a significantly higher AE count rate was recorded at R = 0.125 compared to R = 0.5 when the maximum load was kept the same. This means that the R-ratio would affect the total amount of AE activities collected. It was also found that the dominant frequency range of the AE waveforms directly related to crack growth was 152-487 kHz, and the ranges due to crack closure were likely to be 310 kHz-316 kHz and 500-700 kHz. Based on the proposed frequency ranges, waveform selection was conducted and a better linear relationship between count rate and crack growth rate was observed. This study provides a better understanding of the AE sources and waveforms for future structural health monitoring applications.

Performance Comparison of Multiple Convolutional Neural Networks for Concrete Defects Classification.

Periodical vision-based inspection is a principal form of structural health monitoring (SHM) technique. Over the last decades, vision-based artificial intelligence (AI) has successfully facilitated an effortless inspection system owing to its exceptional ability of accuracy of defects' pattern recognition. However, most deep learning (DL)-based methods detect one specific type of defect, whereas DL has a high proficiency in multiple object detection. This study developed a dataset of two types of defects, i.e., concrete crack and spalling, and applied various pre-built convolutional neural network (CNN) models, i.e., VGG-19, ResNet-50, InceptionV3, Xception, and MobileNetV2 to classify these concrete defects. The dataset developed for this study has one of the largest collections of original images of concrete crack and spalling and avoided the augmentation process to replicate a more real-world condition, which makes the dataset one of a kind. Moreover, a detailed sensitivity analysis of hyper-parameters (i.e., optimizers, learning rate) was conducted to compare the classification models' performance and identify the optimal image classification condition for the best-performed CNN model. After analyzing all the models, InceptionV3 outperformed all the other models with an accuracy of 91%, precision of 83%, and recall of 100%. The InceptionV3 model performed best with optimizer stochastic gradient descent (SGD) and a learning rate of 0.001.