Health Monitoring Of Infrastructure Research Articles

A Cascade Broad Neural Network for Concrete Structural Crack Damage Automated Classification

Crack is the earlier indication of concrete structural severe damage; it plays an important role in structure health monitoring (SHM) of industrial civil infrastructures (such as buildings, bridges, roads, dams, etc.). Crack damage classification is the first and critical stage for concrete SHM. However, commonly used human visual classification is costly, labor-intensive, and unreliable, other machine learning based classification methods also have some drawbacks. To address these problems, this article proposes a cascade broad neural network architecture for concrete surface structural crack damage automated classification, which generates an effective and efficient framework with much less hyper-parameters than deep neural networks, and sufficiently explores the advantages of multilevel cascades of classifier ensemble. Experimental results on four challenging datasets demonstrate that its performance is quite more excellent than current mainstream classification methods (both in testing accuracy and training time).

Strain-Detecting properties of hybrid PE and steel fibers reinforced cement composite (Hy-FRCC) with Multi-Walled carbon nanotube (MWCNT) under repeated compression

This experimental research investigates the feasibility of using hybrid polyethylene (PE) and steel fibers reinforced cement composite (Hy-FRCC) with multi-walled carbon nanotube (MWCNT) as strain sensors. Smart construction material presented in the study is a new type of multifunctional cement composite with both structural material and self-sensing functionalities. The effect of MWCNT dosage (1 and 2% by weight of cement) and the magnitude of repeated compressive stress on the strain-detecting function and elastic modulus was studied. The history of applied compressive loading consists of five successive levels (30, 50, 60, 70 and 80% of compressive strength) of stress amplitude to evaluate the repeated strain-sensing ability of Hy-FRCC. Test results revealed that the incorporation of MWCNT within 2% has little effect on the compressive strength of Hy-FRCC at curing ages of 28-day while under repeated compression, MWCNT is effective to decrease the reduction of the elastic modulus in the Hy-FRCC. Piezoresistivity experiments indicated that the change in resistivity of Hy-FRCC-2 (with 2% MWCNT) specimen can simulate the changes in the measured strain during repeated compressive loading at increasing stress amplitudes up to 80% of compressive strength. It is found that the Hy-FRCC-2 mixture proposed in the present study is likely to be used as strain-sensors for structural health monitoring (SHM) of civil infrastructures.

Existing infrastructure cost effective informative modelling with multisource sensed data: TLS, MMS and photogrammetry

Current practice in large infrastructure management requires new procedures for cost-effective maintenance and safe operation. In this sense, the availability of new surveying, inspection and monitoring procedures as well as new instruments can play a significant role in securing a fundamental asset with an affordable cost. Indeed, inspection and monitoring with automated or semi-automated instruments can guarantee a significant reduction of time in operations that are currently manual, risky and time-consuming. The availability of new sensing platforms and instruments like lightweight mobile mapping systems and drones is fundamental for rethinking current survey and inspection practice. However, their performance, in terms of metric accuracy and reliability, in real applications is still limited. On the other hand, the acquisition of primary data is only the first step of the maintenance workflow. Indeed, even if the adoption of informative content models for structural health monitoring (SHM) for large infrastructures clearly presents important advantages compared with standard management; the adoption of proper tolls for such complex infrastructures poses some issues that need to be solved to develop smooth management and maintenance workflows. This paper presents a methodology for the generation of a detailed informative model starting from multi-sources data: terrestrial laser scanning (TLS), Mobile Mapping Systems (MMS) and photogrammetry. The integration of those different techniques is discussed and a comparative analysis is carried out. The enrichment of the informative model with further data coming from load testing and inspection is presented to create a unique informative platform suitable for different end-users involved in maintenance operations.

A validated robust and automatic procedure for vibration analysis of bridge structures using MEMS accelerometers

Abstract Today, short- and long-term structural health monitoring (SHM) of bridge infrastructures and their safe, reliable and cost-effective maintenance has received considerable attention. From a surveying or civil engineer’s point of view, vibration-based SHM can be conducted by inspecting the changes in the global dynamic behaviour of a structure, such as natural frequencies (i. e. eigenfrequencies), mode shapes (i. e. eigenforms) and modal damping, which are known as modal parameters. This research work aims to propose a robust and automatic vibration analysis procedure that is so-called robust time domain modal parameter identification (RT-MPI) technique. It is novel in the sense of automatic and reliable identification of initial eigenfrequencies even closely spaced ones as well as robustly and accurately estimating the modal parameters of a bridge structure using low numbers of cost-effective micro-electro-mechanical systems (MEMS) accelerometers. To estimate amplitude, frequency, phase shift and damping ratio coefficients, an observation model consisting of: (1) a damped harmonic oscillation model, (2) an autoregressive model of coloured measurement noise and (3) a stochastic model in the form of the heavy-tailed family of scaled t-distributions is employed and jointly adjusted by means of a generalised expectation maximisation algorithm. Multiple MEMS as part of a geo-sensor network were mounted at different positions of a bridge structure which is precalculated by means of a finite element model (FEM) analysis. At the end, the estimated eigenfrequencies and eigenforms are compared and validated by the estimated parameters obtained from acceleration measurements of high-end accelerometers of type PCB ICP quartz, velocity measurements from a geophone and the FEM analysis. Additionally, the estimated eigenfrequencies and modal damping are compared with a well-known covariance driven stochastic subspace identification approach, which reveals the superiority of our proposed approach. We performed an experiment in two case studies with simulated data and real applications of a footbridge structure and a synthetic bridge. The results show that MEMS accelerometers are suitable for detecting all occurring eigenfrequencies depending on a sampling frequency specified. Moreover, the vibration analysis procedure demonstrates that amplitudes can be estimated in submillimetre range accuracy, frequencies with an accuracy better than 0.1 Hz and damping ratio coefficients with an accuracy better than 0.1 and 0.2 % for modal and system damping, respectively.

Toward data anomaly detection for automated structural health monitoring: Exploiting generative adversarial nets and autoencoders

Damage detection is one of the most important tasks for structural health monitoring of civil infrastructure. Before a damage detection algorithm can be applied, the integrity of the data must be ensured; otherwise results may be misleading or incorrect. Indeed, sensor system malfunction, which results in anomalous data (often called faulty data), is a serious problem, as the sensors usually must operate in extremely harsh environments. Identifying and eliminating anomalies in the data is crucial to ensuring that reliable monitoring results can be achieved. Because of the vast amounts of data typically collected by a structural health monitoring system, manual removal of the anomalous data is prohibitive. Machine learning methods have the potential to automate the process of data anomaly detection. Although supervised methods have been proven to be effective for detecting data anomalies, two unresolved challenges reduce the accuracy of anomaly detection: (1) the class imbalance and (2) incompleteness of anomalous patterns of training dataset. Unsupervised methods have the potential to address these challenges, but improvements are required to deal with vast amounts of monitoring data. In this article, the generative adversarial networks are combined with a widely applied unsupervised method, that is, autoencoders, to improve the performance of existing unsupervised learning methods. In addition, the time-series data are transformed to Gramian Angular Field images so that advanced computer vision methods can be included in the network. Two structural health monitoring datasets from a full-scale bridge, including examples of anomalous data caused by sensor system malfunctions, are utilized to validate the proposed methodology. Results show that the proposed methodology can successfully identify data anomalies with good accuracy and robustness, hence can overcome one of the key difficulties in achieving automated structural health monitoring.

Effect of polyvinylidene fluoride on the acoustic impedance matching, poling enhancement and piezoelectric properties of 0–3 smart lead-free piezoelectric Portland cement composites

Smart composites, which are composed of piezoelectric ceramics and cement-based materials, have been developed for sensing and health monitoring of concrete infrastructure. Barium titanate (BT)-Portland cement (PC)-polyvinylidene fluoride (PVDF) composites with 40-60 vol.% BT particles (with a median size of ~ 425 µm) and 0-7 vol.% PVDF were fabricated by the pressing and curing method. The effects of the PVDF content on the acoustic impedance (Zc), dielectric constant (er), piezoelectric charge coefficient (d33) and piezoelectric voltage coefficient (g33) of the BT-PC-PVDF composites were investigated. Increasing the PVDF content in the composites led to a decrease in the Zc value so that it approached that of concrete. The PC phase was replaced by the PVDF phase in the BT-cement-based composites to reduce the leakage current in these composites. As a result, the piezoelectric properties of the BT-PC-PVDF composites improved noticeably. BT contents of 50 and 60 vol.% and a PVDF of 5 vol.% had the highest d33~24-25 pC/N. Moreover, the composite with 50 vol.% BT and 5 vol.% PVDF had the highest piezoelectric voltage coefficient herein (g33=14.6 × 10-3 V∙m/N).

Health Monitoring of Civil Infrastructures by Subspace System Identification Method: An Overview

Structural health monitoring (SHM) is the main contributor of the future’s smart city to deal with the need for safety, lower maintenance costs, and reliable condition assessment of structures. Among the algorithms used for SHM to identify the system parameters of structures, subspace system identification (SSI) is a reliable method in the time-domain that takes advantages of using extended observability matrices. Considerable numbers of studies have specifically concentrated on practical applications of SSI in recent years. To the best of author’s knowledge, no study has been undertaken to review and investigate the application of SSI in the monitoring of civil engineering structures. This paper aims to review studies that have used the SSI algorithm for the damage identification and modal analysis of structures. The fundamental focus is on data-driven and covariance-driven SSI algorithms. In this review, we consider the subspace algorithm to resolve the problem of a real-world application for SHM. With regard to performance, a comparison between SSI and other methods is provided in order to investigate its advantages and disadvantages. The applied methods of SHM in civil engineering structures are categorized into three classes, from simple one-dimensional (1D) to very complex structures, and the detectability of the SSI for different damage scenarios are reported. Finally, the available software incorporating SSI as their system identification technique are investigated.

Imaging techniques for defect detection of fiber reinforced polymer‐bonded civil infrastructures

The present article provides a state-of-the-art review of imaging techniques used for defect detection of fiber reinforced polymer (FRP)-bonded civil engineering structures. Compared to the conventional techniques by equipping a structure with stress wave sensors, the contactless imaging techniques feature efficient instrumentation, convenient data acquisition, and rapid evaluation in the procedures of nondestructive testing (NDT). Recently, a substantial progress in utilizing electromagnetic waves for development of imaging NDT techniques (e.g., synthetic aperture radar, infrared thermography, laser shearography, and laser reflection technique) with the purpose to identify debonding or delamination of FRP-bonded structural systems has been made. As another electromagnetic wave-based imaging technique, X-ray computed tomography is promising for exploration of the damage evolution in this structural system, despite little application in structural health monitoring of real infrastructures. Apart from these imaging techniques, more recently, there have been computer-aided motion magnification techniques for defect identification. The motion magnification technique only requires a digital camera and a computer with vision algorithm, which can amplify the motion of defect region and render it visible in a video. This advanced imaging technique achieves high-resolution measurement, simultaneous full-field inspection, and straightforward defect identification in the scene. In this review, both Eulerian motion magnification and phase-based motion magnification techniques for structural visual inspection are presented and discussed. Furthermore, the present article recommends the combination of imaging NDT techniques with artificial intelligence approaches (e.g., deep learning algorithms) to realize the automated and efficient defect detection towards FRP-bonded civil infrastructures.

Structural modal identification and health monitoring of building structures using self-sensing cementitious composites

Recently self-sensing cementitious composite has demonstrated its strong potentiality for structural health monitoring of civil infrastructures because of its low-cost, long-term stability and compatibility with concrete structures. In this paper, we propose novel hybrid nanocarbon materials engineered cement-based sensors (HNCSs) with high-sensitivity, which are fabricated with self-sensing cementitious composites containing electrostatic self-assembled CNT/NCB composite fillers. The mechanical property and sensing performance of the HNCSs are pre-characterized under static and dynamic compressive loadings. The HNCSs are then integrated into a five-story building model via custom-made clamps to verify the feasibility for dynamic response measurements. Results show that the developed sensors have satisfactory mechanical property and excellent pressure-sensitive reproducibility and stability. With clamps holding on the building model, the HNCSs perform satisfactorily under sinusoidal excitations in the frequency range from 2 to 40 Hz. In addition, the modal frequencies and their changes of the building model caused by ‘damage’ simulated through adding additional masses identified by the HNCSs are favorably consistent with the counterparts acquired by accelerometers and strain gauges, indicating that the developed HNCSs have great potential for structural modal identification and damage detection applications.

Quantifying the Relative Thickness of Conductive Ferromagnetic Materials Using Detector Coil-Based Pulsed Eddy Current Sensors.

Thickness quantification of conductive ferromagnetic materials by means of non-destructive evaluation (NDE) is a crucial component of structural health monitoring of infrastructure, especially for assessing the condition of large diameter conductive ferromagnetic pipes found in the energy, water, oil, and gas sectors. Pulsed eddy current (PEC) sensing, especially detector coil-based PEC sensor architecture, has established itself over the years as an effective means for serving this purpose. Approaches for designing PEC sensors as well as processing signals have been presented in previous works. In recent years, the use of the decay rate of the detector coil-based time domain PEC signal for the purpose of thickness quantification has been studied. Such works have established that the decay rate-based method holds generality to the detector coil-based sensor architecture, with a degree of immunity to factors such as sensor shape and size, number of coil turns, and excitation current. Moreover, this method has shown its effectiveness in NDE of large pipes made of grey cast iron. Following such literature, the focus of this work is explicitly PEC sensor detector coil voltage decay rate-based conductive ferromagnetic material thickness quantification. However, the challenge faced by this method is the difficulty of calibration, especially when it comes to applications such as in situ pipe condition assessment since measuring electrical and magnetic properties of certain pipe materials or obtaining calibration samples is difficult in practice. Motivated by that challenge, in contrast to estimating actual thickness as done by some previous works, this work presents a protocol for using the decay rate-based method to quantify relative thickness (i.e., thickness of a particular location with respect to a maximum thickness), without the requirement for calibration.

Efficient Payload Compression in IP-based Wireless Sensor Network: Algorithmic Analysis and Implementation

Due to its efficiency in end-to-end communication, wireless sensor network based on the Internet protocol (IP-WSN) is used for monitoring purposes. Nowadays, the concerned agencies are giving their highest priority to monitor its critical infrastructure. Infrastructure health monitoring is the measure of estimating the state of infrastructure health or detecting the changes in structures that affect its performance. The traditional approach to monitor the infrastructure health is done by using centralized data acquisition hub. Installation and commissioning of these systems represent significant concerns, thus moving toward IP-WSN. As cost effectiveness and energy efficiency are major concerns, our proposed approach is to reduce the amount of overhead while keeping the infrastructure health monitoring system accurate. Our contribution in this paper is to reduce the amount of data to be transmitted by compressing the payload of the packets. Thus, we have proposed a double compression algorithm. In this way, the capacity of the sensor node will be increased since less time will be taken to transmit data between the intermediate node as well as the coordinator node. As a consequence, it will also extend the lifetime of the battery.

RFID-Based Wireless Multi-Sensory System for Simultaneous Dynamic Acceleration and Strain Measurements of Civil Infrastructure

In this paper, we develop a radio frequency identification (RFID)-based wireless multi-sensory infrastructure health monitoring (IHM) system that can simultaneously measure dynamic acceleration and strain. The system consists of a novel multi-sensor integrated semi-passive ultra-high frequency (UHF) tag antenna that can be mounted on civil infrastructure elements; even made out of metal. The system is capable of measuring 3-axis dynamic acceleration and strain with spectral bandwidths of 40 Hz and 26.5 Hz, respectively. The natural frequency determination of infrastructure by the dynamic acceleration and strain measurements of the proposed system is accurate to 60 mHz. Benchmarking of the RFID-based wireless multi-sensory system is provided by comprehensive comparison of the results with measurements from a commercial wireless strain measurement system. The proposed system has 30 mHz natural frequency determination error when compared with dynamic strain measurement from the commercial system.

A Machine Learning Approach for Structural Health Monitoring Using Noisy Data Sets

Continuous structural health monitoring of civil infrastructure can be achieved by deploying an Internet of Things network of distributed acceleration sensors in buildings to capture floor movement. Postdisaster damage levels can be computed based on the peak relative floor displacement as specified in government standards. This article uses machine learning approaches to identify the status of buildings postevent based on accelerometer traces. Prior work in the field assumed the use of high-quality accelerometers for displacement estimation. In this article, we focus on using lower quality and cheaper accelerometers, while accounting for noise effects by incorporating noisy data sets in machine learning approaches for classification. A labeled acceleration data set of buildings response to earthquakes was created, where each sample is labeled with its corresponding damage severity. Sensor noise is included in the data set to model nonideal sensors. Classification performance of machine learning algorithms, such as support vector machine, K-nearest neighbor, and convolutional neural network, is presented. Techniques for addressing noise levels are proposed, and the results are compared with regular noise cancellation techniques that adopt high-pass filtering. Note to Practitioners —This article presents a methodology for automatic estimation of buildings status in the aftermath of a natural disaster, such as an earthquake. It focuses on using low-cost inertial sensors, such as accelerometers, to sense buildings’ vibrations and then applying machine learning algorithms to detect damage. Utilizing the convolutional network approach, the proposed methods detect the building damage state with high accuracy. Since this article focuses on using cheap sensors, the cost of deploying a sensor network to monitor buildings is reduced significantly. Deploying this network enables rescue and reconnaissance teams to have a clear view of the most vulnerable structures.

Presenting a New Wireless Strain Method for Structural Monitoring: Experimental Validation

The structural health monitoring (SHM) of large and complex infrastructures as well as laboratory tests of new structures and materials resorts to strain gauge measurements to check mechanical stress. A wireless measurement of the strain gauge response is desirable in many practical applications to avoid the cost and the difficulty of wiring, particularly in large structures requiring several sensors and in complex objects where the measurement points are difficult to access. In this paper, a wireless strain gauge which is a hybrid between an RFID tag and a usual thin-film resistive strain gauge is experimented. Installation and maintenance problems of the wireless sensor networks are overcome allowing a high level of measurement accuracy, comparable to that of wired strain sensors, together with a long measurement distance. A large set of measurements has been performed using reference specimens and readings in order to validate the sensor and to develop a calibration procedure that makes the sensor suitable for a large number of different applications in civil engineering.

A review on deep learning-based structural health monitoring of civil infrastructures

In the past two decades, structural health monitoring (SHM) systems have been widely installed on various civil infrastructures for the tracking of the state of their structural health and the detection of structural damage or abnormality, through long-term monitoring of environmental conditions as well as structural loadings and responses. In an SHM system, there are plenty of sensors to acquire a huge number of monitoring data, which can factually reflect the in-service condition of the target structure. In order to bridge the gap between SHM and structural maintenance and management (SMM), it is necessary to employ advanced data processing methods to convert the original multi-source heterogeneous field monitoring data into different types of specific physical indicators in order to make effective decisions regarding inspection, maintenance and management. Conventional approaches to data analysis are confronted with challenges from environmental noise, the volume of measurement data, the complexity of computation, etc., and they severely constrain the pervasive application of SHM technology. In recent years, with the rapid progress of computing hardware and image acquisition equipment, the deep learning-based data processing approach offers a new channel for excavating the massive data from an SHM system, towards autonomous, accurate and robust processing of the monitoring data. Many researchers from the SHM community have made efforts to explore the applications of deep learning-based approaches for structural damage detection and structural condition assessment. This paper gives a review on the deep learning-based SHM of civil infrastructures with the main content, including a brief summary of the history of the development of deep learning, the applications of deep learning-based data processing approaches in the SHM of many kinds of civil infrastructures, and the key challenges and future trends of the strategy of deep learning-based SHM.

Corrosion Sensors for Structural Health Monitoring of Oil and Natural Gas Infrastructure: A Review

Corrosion has been a great concern in the oil and natural gas industry costing billions of dollars annually in the U.S. The ability to monitor corrosion online before structural integrity is compromised can have a significant impact on preventing catastrophic events resulting from corrosion. This article critically reviews conventional corrosion sensors and emerging sensor technologies in terms of sensing principles, sensor designs, advantages, and limitations. Conventional corrosion sensors encompass corrosion coupons, electrical resistance probes, electrochemical sensors, ultrasonic testing sensors, magnetic flux leakage sensors, electromagnetic sensors, and in-line inspection tools. Emerging sensor technologies highlight optical fiber sensors (point, quasi-distributed, distributed) and passive wireless sensors such as passive radio-frequency identification sensors and surface acoustic wave sensors. Emerging sensors show great potential in continuous real-time in-situ monitoring of oil and natural gas infrastructure. Distributed chemical sensing is emphasized based on recent studies as a promising method to detect early corrosion onset and monitor corrosive environments for corrosion mitigation management. Additionally, challenges are discussed including durability and stability in extreme and harsh conditions such as high temperature high pressure in subsurface wellbores.

Evaluation of UAV–CRP Data for Monitoring Transportation Infrastructure Constructed over Expansive Soils

Application of unmanned aerial vehicles (UAVs) for civil infrastructure monitoring has gained impetus owing to the advancement of aerial platforms paralleled with the development of sophisticated sensors. Photogrammetry is the science of measuring distances from two or more images, and close-range photogrammetry (CRP) is a part of photogrammetry that involves calculating measurements of an object within a maximum distance of 305 m away from the inspecting sensors. Geotechnical problems including differential heaving and related cracking of expansive soils cause extensive damage to pavement infrastructure. Feasibility of using UAV–CRP technology in health monitoring of pavement infrastructure constructed over problematic soils has been comprehensively studied, and this paper presents an overview of these results. Pavement performance data including longitudinal and transverse slopes, as well as distress conditions such as pavement surface cracking, pothole formation, and rutting or excessive deformation, are monitored via UAV–CRP technology, and these data sets are comprehensively analysed. UAV–CRP-interpreted performance indicators showed a very good agreement with those obtained from traditional methods surveys and profiler studies. With the further research, the UAV–CRP technology will play an important engineering role in safe, inexpensive and comprehensive health monitoring of infrastructure built over problematic soil conditions.

Application of Machine Learning with Impedance Based Techniques for Structural Health Monitoring of Civil Infrastructure

Increased attentiveness on the environmental and effects of aging, deterioration and extreme events on civil infrastructure has created the need for more advanced damage detection tools and structural health monitoring (SHM). Today, these tasks are performed by signal processing, visual inspection techniques along with traditional well known impedance based health monitoring EMI technique. New research areas have been explored that improves damage detection at incipient stage and when the damage is substantial. Addressing these issues at early age prevents catastrophe situation for the safety of human lives. To improve the existing damage detection newly developed techniques in conjugation with EMI innovative new sensors, signal processing and soft computing techniques are discussed in details this paper. The advanced techniques (soft computing, signal processing, visual based, embedded IOT) are employed as a global method in prediction, to identify, locate, optimize, the damage area and deterioration. The amount and severity, multiple cracks on civil infrastructure like concrete and RC structures (beams and bridges) using above techniques along with EMI technique and use of PZT transducer. In addition to survey advanced innovative signal processing, machine learning techniques civil infrastructure connected to IOT that can make infrastructure smart and increases its efficiency that is aimed at socioeconomic, environmental and sustainable development.

Non-destructive Assessment and Health Monitoring of Railway Infrastructures

A continuous increase in the worldwide demand for high-speed traffic, freight tonnage as well as of the train operating frequency is worsening the decay conditions of many railway infrastructures. This occurrence affects economy-related business as well as contributing to rising maintenance costs. It is known that a failure of a railway track may result in tremendous economic losses, legal liabilities, service interruptions and, eventually, fatalities. Parallel to this, requirements to maintain acceptable operational standards are very demanding. In addition to the above, a main issue nowadays in railway engineering is a general lack of funds to allow safety and comfort of the operations as well as a proper maintenance regime of the infrastructures. This is mostly the result of a traditional approach that, on average, tends to invest in high-priority costs, such as safety-related costs, compromising lower-priority interventions (e.g., quality and comfort of the operations). A solution to correct this trend can be moving from a reactive to a proactive action planning approach in order to limit more effectively the likelihood of progressive rail track decay. Within this context, this paper reports a review on the use of traditional and non-destructive testing (NDT) methods for the assessment and health monitoring of railway infrastructures. State-of-the-art research on a stand-alone use of NDT methods or a combination of them for quality control, inspection and maintenance tasks in this subject area is discussed.

Analysis of a Decentralized Mobility Management Support for Adaptive Structural Health Monitoring of SOC Public Infrastructures

Analysis of a Decentralized Mobility Management Support for Adaptive Structural Health Monitoring of SOC Public Infrastructures