Search

The correct workability of concrete is an essential parameter for its placement and compaction. However, an absence of automatic and transparent measurement methods to estimate the workability of concrete hinders the adaptation from laborious traditional methods such as the slump test. In this paper, we developed a machine-learning framework for estimating the slump class of concrete in the mixer using a stereovision camera. Depth data from five different slump classes was transformed into Haralick texture features to train several machine-learning classifiers. The best-performing classifier achieved a multiclass classification accuracy of 0.8179 with the XGBoost algorithm. Furthermore, we found through statistical analysis that while the denoising of depth data has little effect on the accuracy, the feature extraction of mixer blades and the choice of region of interest significantly increase the accuracy and the efficiency of the classifiers. The proposed framework shows robust results, indicating that stereovision is a competitive solution to estimate the workability of concrete during concrete production. [ABSTRACT FROM AUTHOR]

The moisture content within the concrete pore network significantly influences the mechanical, thermal, and durability characteristics of concrete structures. This paper introduces a novel fully embedded wireless temperature and relative humidity sensor connected to an automatic acquisition system designed for continuous concrete monitoring. Relative humidity measurements from this new sensor are compared with those obtained by a commercial system based on the borehole method at different depths (2.5 and 4.0 cm) and exposure conditions (oven drying and humid chamber). The results allow for proving that both systems provide consistent internal relative humidity measurements aligned with the exposure conditions and highlight the capability of fully embedded wireless sensors as a practical and reliable alternative to the conventional borehole method. Additionally, the continuous monitoring of the wireless cast-in sensor exhibits reliability during unintended temperature fluctuations, emphasizing the effectiveness of permanently installed sensors in promptly detecting unintended curing variations in real time. The continuous real-time information provided combined with the practicality of these sensors might assist construction managers to improve the quality control of the concrete curing process and shrinkage behavior, and ensure the integrity of concrete surface finishing. [ABSTRACT FROM AUTHOR]

Coarse aggregate in concrete is basically free from sulfate corrosion. If the influence of the coarse aggregate in the concrete is not eliminated, the change amount of the concrete ultrasonic pulse velocity value is directly used to evaluate the damage degree of sulfate corrosion in the concrete, and the results are often inaccurate. This paper presents an evaluation method of corrosion damage for the sulfate-attacked concrete by CT, ultrasonic velocity testing and AHP methods. CT was used to extract the coarse aggregate information in the specimen, and the proportion of coarse aggregate on the ultrasonic test line was calculated based on CT image analysis. Then, the correction value of ultrasonic pulse velocity (UPV) of the concrete structure was calculated, and the sulfate corrosion degree of concrete structure was evaluated using the analytic hierarchy process (AHP). The results show that the evaluation method proposed in this paper could more accurately evaluate the corrosion damage in the sulfate-attacked concrete structures, and the evaluation results were more in line with reality. [ABSTRACT FROM AUTHOR]

This paper aims to investigate wave dispersion behavior in the quasi-solid state of concrete to better understand microstructure hydration interactions. The quasi-solid state refers to the consistency of the mixture between the initial liquid–solid stage and the hardened stage, where the concrete has not yet fully solidified but still exhibits viscous behavior. The study seeks to enable a more accurate evaluation of the optimal time for the quasi-liquid product of concrete using both contact and noncontact sensors, as current set time measurement approaches based on group velocity may not provide a comprehensive understanding of the hydration phenomenon. To achieve this goal, the wave dispersion behavior of P-wave and surface wave with transducers and sensors is studied. The dispersion behavior with different concrete mixtures and the phase velocity comparison of dispersion behavior are investigated. The analytical solutions are used to validate the measured data. The laboratory test specimen with w/c = 0.5 was subjected to an impulse in a frequency range of 40 kHz to 150 kHz. The results demonstrate that the P-wave results exhibit well-fitted waveform trends with analytical solutions, showing a maximum phase velocity when the impulse frequency is at 50 kHz. The surface wave phase velocity shows distinct patterns at different scanning times, which is attributed to the effect of the microstructure on the wave dispersion behavior. This investigation delivers profound knowledge of hydration and quality control in the quasi-solid state of concrete with wave dispersion behavior, providing a new approach for determining the optimal time of the quasi-liquid product. The criteria and methods developed in this paper can be applied to optimal timing for additive manufacturing of concrete material for 3D printers by utilizing sensors. [ABSTRACT FROM AUTHOR]

Concrete structures have emerged as some of the most extensively utilized materials in the construction industry due to their inherent plasticity and high-strength characteristics. However, due to the temperature fluctuations, humidity, and damage caused by human activities, challenges such as crack propagation and structural failures pose threats to the safety of people's lives and property. Meanwhile, conventional non-destructive testing methods are limited to defect detection and lack the capability to provide real-time monitoring and evaluating of concrete structural stability. Consequently, there is a growing emphasis on the development of effective techniques for monitoring the health of concrete structures, facilitating prompt repairs and mitigation of potential instabilities. This paper comprehensively presents traditional and novel methods for concrete structural properties and damage evolution monitoring, including emission techniques, electrical resistivity monitoring, electromagnetic radiation method, piezoelectric transducers, ultrasonic techniques, and the infrared thermography approach. Moreover, the fundamental principles, advantages, limitations, similarities and differences of each monitoring technique are extensively discussed, along with future research directions. Each method has its suitable monitoring scenarios, and in practical applications, several methods are often combined to achieve better monitoring results. The outcomes of this research provide valuable technical insights for future studies and advancements in the field of concrete structural health monitoring. [ABSTRACT FROM AUTHOR]

This paper presents a ceramic stress sensor with the dimension of a coin, able to measure the compressive force (stress) applied to its two round faces. The sensor is designed and engineered to be embedded inside concrete or masonry structures, like bridges or buildings. It provides good accuracy, robustness, and simplicity of use at potentially low cost for large-scale applications in civil structures. Moreover, it can be calibrated temperature compensated, and it is inherently hermetic, ensuring the protection of sensitive elements from the external environment. It is, therefore, suitable for operating in harsh and dirty environments like civil constructions. The sensor directly measures the internal stress of the structure, exploiting the piezo resistivity of thick film ink based on ruthenium oxide. It is insensitive with respect to the stiffness of the embedding material and the variation of the surrounding material properties like concrete hardening, shrinkage, and creep as it decouples the two components of stress. [ABSTRACT FROM AUTHOR]

In this paper, we introduce a method for simulating the deformation of concrete surfaces due to weathering employing an example-based approach to replicate shape changes observed in real-world objects. A key challenge in implementing this approach is the scarcity of opportunities to measure shapes both before and after the weathering process. To overcome this limitation, we utilize concrete bricks collected from real-world environments as standardized examples, allowing for an analysis of erosion. By measuring erosion based on the estimated original shape, we correlate the characteristics of erosion with geometric features such as curvature and accessibility. We then apply this analysis to simulate new weathering effects in a given input model in alignment with its own geometric features. Our method yields visually compelling results while reproducing the variation of geometric weathering effects. [ABSTRACT FROM AUTHOR]

Modern infrastructure heavily relies on robust concrete structures, underscoring the critical need for effective monitoring to ensure their safety and durability. This paper addresses this imperative issue by introducing an innovative automated and wireless system for continuous structural monitoring. By employing embedded electrical resistivity sensors coupled with a wireless-based data transmission mechanism, real-time data collection becomes feasible. We provide a general description of the system's architecture and its application in a pilot study covering the effects of the devices on concrete properties and data transmission. The dielectric properties of concrete specimens were investigated under natural and accelerated curing/degradation and the results were used in the final design of the antenna device. Furthermore, a pilot test comprising four reinforced concrete columns was used to investigate the range of data transmission from inside to outside of the concrete, the effects of the hardware device on the compressive strength and concrete distribution in the columns, and the data transmission quality in real time under realistic exposure conditions. [ABSTRACT FROM AUTHOR]

In the field of non-destructive testing (NDT) for concrete structures, the traditional air-coupled impact-echo technology often has the problems of complex operation and low efficiency. In order to solve these problems, this study uses Comsol software to establish a finite element model (FEM) of the concrete structure with different void sizes and obtains the variation rule of peak frequency. The recognition property of the concrete void based on peak frequency is proposed, which is explained and validated by relevant theory and experiments. The results show that compared with the depth of the void, the influence of the void width on the peak frequency increases significantly. When the void width is greater than 0.3 m, the peak frequency of the sound wave decreases with the increase in the width, and the change is obvious. This paper describes the applicability of concrete void depth less than 0.4 m for the air-coupled method and, when the concrete void depth is less than 0.4 m, the peak frequency can be used to effectively identify void widths greater than 0.3 m. The research results will be beneficial to void detection of concrete structures such as tunnel lining and pavements. [ABSTRACT FROM AUTHOR]

This paper proposes a new intelligent recognition method for concrete ultrasonic detection based on wavelet packet transform and a convolutional neural network (CNN). To validate the proposed data-based method, a case study is presented where the K-fold cross-validation was adopted to produce the performance analysis and classification experiments. Moreover, three evaluation indicators, precision, recall, and F-score, are calculated for analyzing the classification performance of the trained models. As a result, the obtained four-classifying CNN reaches more than 99% detection accuracy while the lowest recognition accuracy is not less than 92.5% on the testing dataset for the six-classifying CNN model. Compared with the existing stochastic configuration network (SCN) models, the presented method achieves the design objective with better recognition performance. The calculation results of the six-classifying and five-classifying models and related research clearly indicate the remaining challenging tasks for intelligent recognition algorithms in extracting features and classifying mass data from various concrete defects precisely and efficiently. [ABSTRACT FROM AUTHOR]

This paper proposes a new technique for the construction of a concrete-beam health indicator based on the Kullback–Leibler divergence (KLD) and deep learning. Health indicator (HI) construction is a vital part of remaining useful lifetime (RUL) approaches for monitoring the health of concrete structures. Through the construction of a HI, the deterioration process can be processed and portrayed so that it can be forwarded to a prediction module for RUL prognosis. The degradation progression and failure can be identified by predicting the RUL based on the situation of the current specimen; as a result, maintenance can be planned to reduce safety risks, reduce financial costs, and prolong the specimen's useful lifetime. The portrayal of deterioration through HI construction from raw acoustic emission (AE) data is performed using a deep neural network (DNN), whose parameters are obtained by pretraining and fine tuning using a stack autoencoder (SAE). Kullback–Leibler divergence, which is calculated between a reference normal-conditioned signal and a current unknown signal, was used to represent the deterioration process of concrete structures, which has not been investigated for the concrete beams so far. The DNN-based constructor then learns to generate HI from raw data with KLD values as the training label. The HI construction result was evaluated with run-to-fail test data of concrete specimens with two measurements: fitness analysis of the construction result and RUL prognosis. The results confirm the reliability of KLD in portraying the deterioration process, showing a large improvement in comparison to other methods. In addition, this method requires no adept knowledge of the nature of the AE or the system fault, which is more favorable than model-based approaches where this level of expertise is compulsory. Furthermore, AE offers in-service monitoring, allowing the RUL prognosis task to be performed without disrupting the specimen's work. [ABSTRACT FROM AUTHOR]

Buildings and infrastructure in congested metropolitan areas are continuously deteriorating. Various structural flaws such as surface cracks, spalling, delamination, and other defects are found, and keep on progressing. Traditionally, the assessment and inspection is conducted by humans; however, due to human physiology, the assessment limits the accuracy of image evaluation, making it more subjective rather than objective. Thus, in this study, a multivariant defect recognition technique was developed to efficiently assess the various structural health issues of concrete. The image dataset used was comprised of 3650 different types of concrete defects, including surface cracks, delamination, spalling, and non-crack concretes. The proposed scheme of this paper is the development of an automated image-based concrete condition recognition technique to categorize, not only non-defective concrete into defective concrete, but also multivariant defects such as surface cracks, delamination, and spalling. The developed convolution-based model multivariant defect recognition neural network can recognize different types of defects on concretes. The trained model observed a 98.8% defect detection accuracy. In addition, the proposed system can promote the development of various defect detection and recognition methods, which can accelerate the evaluation of the conditions of existing structures. [ABSTRACT FROM AUTHOR]

In this paper, we studied the possibility of increasing the Brillouin frequency shift (BFS) detection accuracy in distributed fibre-optic sensors by the separate and joint use of different algorithms for finding the spectral maximum: Lorentzian curve fitting (LCF, including the Levenberg–Marquardt (LM) method), the backward correlation technique (BWC) and a machine learning algorithm, the generalized linear model (GLM). The study was carried out on real spectra subjected to the subsequent addition of extreme digital noise. The precision and accuracy of the LM and BWC methods were studied by varying the signal-to-noise ratios (SNRs) and by incorporating the GLM method into the processing steps. It was found that the use of methods in sequence gives a gain in the accuracy of determining the sensor temperature from tenths to several degrees Celsius (or MHz in BFS scale), which is manifested for signal-to-noise ratios within 0 to 20 dB. We have found out that the double processing (BWC + GLM) is more effective for positive SNR values (in dB): it gives a gain in BFS measurement precision near 0.4 °C (428 kHz or 9.3 μ ε ); for BWC + GLM, the difference of precisions between single and double processing for SNRs below 2.6 dB is about 1.5 °C (1.6 MHz or 35 μ ε ). In this case, double processing is more effective for all SNRs. The described technique's potential application in structural health monitoring (SHM) of concrete objects and different areas in metrology and sensing were also discussed. [ABSTRACT FROM AUTHOR]

This paper represents a first attempt to study the feasibility of using shear wave (SW) ultrasonic probes as pump-wave sources in concrete microcrack detection and monitoring by Nonlinear Ultrasonic Coda Wave Interferometry (NCWI). The premise behind our study is that the nonlinear elastic hysteretic behavior at microcracks may depend on their orientation with respect to the stationary wave-field induced by the pump-wave source. In this context, the use of a SW probe as a pump-wave source may induce the nonlinear elastic behavior of microcracks oriented in directions not typically detected by a conventional longitudinal pump-wave source. To date, this premise is hard to address by current experimental and numerical methods, however, the feasibility of using SW probes as a pump-wave source can be experimentally tested. This idea is the main focus of the present work. Under laboratory conditions, we exploit the high sensitivity of the CWI technique to capture the transient weakening behaviour induced by the SW pump-wave source in concrete samples subjected to loading and unloading cycles. Our results show that after reaching a load level of 40 % of the ultimate stress, the material weakening increases as a consequence of microcrack proliferation, which is consistent with previous studies. Despite the lack of exhaustive experimental studies, we believe that our work is the first step in the formulation of strategies that involve an appropriate selection and placement of pump-wave sources to improve the NCWI technique. These improvements may be relevant to convert the NCWI technique into a more suitable non-destructive testing technique for the inspection of microcracking evolution in concrete structures and the assessment of their structural integrity. [ABSTRACT FROM AUTHOR]

Terrestrial laser scanning technology (TLS) is a new technique for quickly getting three-dimensional information. In this paper we research the health assessment of concrete structures with a Finite Element Method (FEM) model based on TLS. The goal focuses on the benefits of 3D TLS in the generation and calibration of FEM models, in order to build a convenient, efficient and intelligent model which can be widely used for the detection and assessment of bridges, buildings, subways and other objects. After comparing the finite element simulation with surface-based measurement data from TLS, the FEM model is determined to be acceptable with an error of less than 5%. The benefit of TLS lies mainly in the possibility of a surface-based validation of results predicted by the FEM model. [ABSTRACT FROM AUTHOR]

Hundreds of people are killed or injured annually in the United States in accidents related to ice formation on roadways and bridge decks. In this paper, a novel embedded sensor system is proposed for the detection of black ice as well as wet, dry, and frozen pavement conditions on roads, runways, and bridges. The proposed sensor works by detecting changes in electrical resistance between two sets of stainless steel poles embedded in the concrete sensor to assess surface and near-surface conditions. A preliminary decision algorithm is developed that utilizes sensor outputs indicating resistance changes and surface temperature. The sensor consists of a 102-mm-diameter, 38-mm-high, concrete cylinder. Laboratory results indicate that the proposed sensor can effectively detect surface ice and wet conditions even in the presence of deicing chlorides and rubber residue. This sensor can further distinguish black ice from ice that may exist within concrete pores. [ABSTRACT FROM AUTHOR]

This paper reports on the advantages and drawbacks of available test methods for the determination of chloride content in cementitious materials in general, and the application of Ag/AgCl chloride sensors in particular. The main factors that affect the reliability of a chloride sensor are presented. The thermodynamic behaviour of silver in the presence or absence of chloride ions is described and kinetic restrictions are addressed. The parameters that can affect the activity of chloride ions in the medium and/or the rate of ion exchange and dissolution/precipitation processes at the sensor's surface are also considered. In this regard, the contribution of morphology and microstructure of the AgCl layer, binding of chloride ions and the compactness of hydration products around the chloride sensor are highlighted. The important parameters for a reliable sensor's response are discussed and the possible causes of inaccuracies are evaluated. [ABSTRACT FROM AUTHOR]

Mass concrete is widely used in large-scale projects, including metro upper cover structures, water conservancy dams, and heavy equipment foundations, among others, necessitating the process of health monitoring in mass concrete construction. The development of reliable and simple strength-monitoring methods for mass concrete is challenging because the inner temperature of mass concrete is high and changes a lot. This study proposes a strength-monitoring approach for mass concrete using barium titanate–bismuth ferrite/polyvinylidene fluoride (BT–BFO/PVDF) nanocomposite piezoelectric sensors, wherein the new sensors are embedded as actuators and sensors in mass concrete. The stress wave generated by the BT–BFO/PVDF piezoelectric sensors is used to monitor the specimen's strength for 28 days. The piezoelectric voltage received by the sensors in mass concrete is analyzed. The experimental results indicate that the signal received by the BT–BFO/PVDF sensors is not easily affected by the internal temperature of mass concrete compared with that of the traditional PVDF piezoelectric sensors. The signal parameters sensitive to concrete strength variation and the change trend of concrete strength are closely related to the piezoelectric voltage. Therefore, the proposed approach using BT–BFO/PVDF nanocomposite piezoelectric sensors is efficient (error < 10%) in mass concrete monitoring. Moreover, the monitoring results do not need temperature compensation. The physical meaning of the obtained strength prediction formula is proposed. An experimental system based on PVDF dynamic strain-sensing characteristics is established. [ABSTRACT FROM AUTHOR]

The development and research of an ultrasonic-based concrete structural health monitoring system encounters a variety of problems, such as demands of decreasing complexity, high accuracy, and extendable system output. Aiming at these requirements, a low-cost extendable system based on FPGA with adjustable system output has been designed, and the performance has been evaluated by different assessment parameters set in this paper. Besides the description of the designed system and the experiments in air medium, the residual similarity and Pearson correlation coefficients of experimental and theoretical data have been used to evaluate the submodules' output. The output performance of the overall system is evaluated by the Pearson correlation coefficient, root-mean-square error (RMSE), and magnitude-squared coherence with 40 experimental data. The maximum, median, minimum, and mean values in three-parameter datasets are analyzed for discussing the working condition of the system. The experimental results show that the system works stably and reliably with tunable frequency and amplitude output. [ABSTRACT FROM AUTHOR]

The protection of concrete structures against corrosion in marine environments has always been a challenge due to the presence of a saline solution--A natural corrosive agent to the concrete paste and steel reinforcements. The concentration of salt is a key parameter influencing the rate of corrosion. In this paper, we propose an optical fiber-based salinity sensor based on bundled multimode plastic optical fiber (POF) as a sensor probe and a concave mirror as a reflector in conjunction with an intensity modulation technique. A refractive index (RI) sensing approach is analytically investigated and the findings are in agreement with the experimental results. A maximum sensitivity of 14,847.486/RIU can be achieved at RI = 1.3525. The proposed technique is suitable for in situ measurement and monitoring of salinity in liquid. [ABSTRACT FROM AUTHOR]

Concrete-filled steel tube (CFST) members have been widely used in civil engineering due to their advanced mechanical properties. However, internal defects such as the concrete core voids and interface debonding in CFST structures are likely to weaken their load-carrying capacity and stiffness, which affects the safety and serviceability. Visualizing the inner defects of the concrete cores in CFST members is a critical requirement and a challenging task due to the obvious difference in the material mechanical parameters of the concrete core and steel tube in CFST members. In this study, a curved ray theory-based travel time tomography (TTT) with a least square iterative linear inversion algorithm is first introduced to quantitatively identify and visualize the sizes and positions of the concrete core voids in CFST members. Secondly, a numerical investigation of the influence of different parameters on the inversion algorithm for the defect imaging of CFST members, including the effects of the model weighting matrix, weighting factor and grid size on the void's imaging quality and accuracy, is carried out. Finally, an experimental study on six CFST specimens with mimicked concrete core void defects is performed in a laboratory and the mimicked defects are visualized. The results demonstrate that TTT can identify the sizes and positions of the concrete core void defects in CFST members efficiently with the use of optimal parameters. [ABSTRACT FROM AUTHOR]

Prestressed concrete (PSC) is one of the most reliable, durable and widely used construction materials, which overcomes the weakness of concrete in tension by the introduction of a prestress force. Smart strands enabling measurement of the prestress force have recently been developed to maintain PSC structures throughout their lifetime. However, the smart strand cannot give a representative indication of the whole prestress force when used in multi-strand systems since each strand sustains a different prestress force. In this paper, the actual distribution of the prestress force in a multi-strand system is examined using elastomagnetic (EM) sensors to develop a method for tracking representative indicators of the prestress force using smart strands. [ABSTRACT FROM AUTHOR]

The nondestructive inspection of concrete structures is indispensable for ensuring the safety and reliability of aging infrastructures. Ultrasonic waves having a frequency of tens of kHz are frequently used to reduce the scattering attenuation due to coarse aggregates. Such low frequencies enable the measurement of the thickness of concrete structures and detection of layer-type defects, such as delamination, whereas it causes a lack of sensitivity to crack-type defects. In this paper, to realize the ultrasonic phased array (PA) imaging of crack-type defects, we fabricated a low-frequency (LF) array transducer with a center frequency of hundreds of kHz. To avoid the crosstalk between piezoelectric elements and dampen the vibration of each element, we adopted soft lead zirconate titanate (soft PZT) with a low mechanical quality factor. Subsequently, we optimized the geometry of each piezoelectric element using a finite element method to generate a short pulse. After validating the design in a fundamental experiment using a single-element transducer, we fabricated a 32-element array transducer with a center frequency of 350 kHz. To show the imaging capability of the LF array transducer, we applied it to a concrete specimen with a delamination. As a result, the PA with the LF array transducer clearly visualized the delamination, which could not be visualized using the PA with a 2.5 MHz array transducer. Furthermore, we applied it to a more challenging defect, a slit, which is sometimes used to simulate crack-type defects. As a result, the PA with the LF array transducer clearly visualized a slit of 1 mm width and 40 mm height in a concrete specimen. Thus, we demonstrated the usefulness of the LF array transducer for inspecting crack-type defects. [ABSTRACT FROM AUTHOR]

Concrete is the most commonly used construction material nowadays. With emerging cutting-edge technologies such as nanomaterials (graphene, carbon nanotubes, etc.), advanced sensing (fiber optics, computer tomography, etc.), and artificial intelligence, concrete can now achieve self-sensing, self-healing, and ultrahigh performance. The concept and functions of smart concrete have thus been partially realized. However, due to the wider application location (coastal areas, cold regions, offshore, and deep ocean scenarios) and changing climate (temperature increase, more CO2 emissions, higher moisture, etc.), durability monitoring (pH, ion penetration, carbonation, corrosion, etc.) becomes an essential component for smart concrete. Fiber optic sensors (FOS) have been widely explored in recent years for concrete durability monitoring due to their advantages of high sensitivity, immunity to harsh environments, small size, and superior sensitivity. The purpose of this review is to summarize FOS development and its application in concrete durability monitoring in recent years. The objectives of this study are to (1) introduce the working principle of FOS, including fiber Bragg grating (FBG), long-period fiber grating (LPFG), surface plasmon resonance (SPR), fluorescence-based sensors, and distributed fiber optic sensors (DFOS); (2) compare the sensitivity, resolution, and application scenarios of each sensor; and (3) discuss the advantages and disadvantages of FOS in concrete durability monitoring. This review is expected to promote technical development and provide potential research paths in the future for FOS in durability monitoring in smart concrete. [ABSTRACT FROM AUTHOR]

Diagnostics and assessment of the structural performance of collectors and tunnels require multi-criteria as well as comprehensive analyses for improving the safety based on acquired measurement data. This paper presents the basic goals for a structural health monitoring system designed based on distributed fiber optic sensors (DFOS). The issue of selecting appropriate sensors enabling correct strain transfer is discussed hereafter, indicating both limitations of layered cables and advantages of sensors with monolithic cross-section design in terms of reliable measurements. The sensor's design determines the operation of the entire monitoring system and the usefulness of the acquired data for the engineering interpretation. The measurements and results obtained due to monolithic DFOS sensors are described hereafter on the example of real engineering structure—the Burakowski concrete collector in Warsaw during its strengthening with glass-fiber reinforced plastic (GRP) panels. [ABSTRACT FROM AUTHOR]

Detailed information about concrete behavior in real structures is an important issue in controlling its performance during its service life, and the use of embedded sensors to obtain desired information such as temperature, onset of the cracking process and evolution of strains, has gained the attention of the building concrete industry. Data obtained using this technology can provide valuable information for decision making about the need for corrective interventions that can ensure the integrity and safety of concrete structures for long period of time. This paper presents a review of the current state-of-the-art of embedded fiber optic sensors used to assess concrete information of a wide range of aspects, comprising: existing alternative technologies, characteristics and advantages, practical applications and future developments. Complementarily, the work presents preliminary results of the use of fiber optic sensors to automatically and continuously perform expansion readings of AAR in concrete elements that facilitate both the storage—with elimination of the usual interruptions for manual readings—and the availability of continuous results of expansion data that are not possible to obtain with usual AAR tests code reading recommendations. [ABSTRACT FROM AUTHOR]

In the wide scenario of heritage documentation and conservation, the multi-scale nature of digital models is able to twin the real object, as well as to store information and record investigation results, in order to detect and analyse deformation and materials deterioration, especially from a structural point of view. The contribution proposes an integrated approach for the generation of an n-D enriched model, also called a digital twin, able to support the interdisciplinary investigation process conducted on the site and following the processing of the collected data. Particularly for 20th Century concrete heritage, an integrated approach is required in order to adapt the more consolidated approaches to a new conception of the spaces, where structure and architecture are often coincident. The research plans to present the documentation process for the halls of Torino Esposizioni (Turin, Italy), built in the mid-twentieth century and designed by Pier Luigi Nervi. The HBIM paradigm is explored and expanded in order to fulfil the multi-source data requirements and adapt the consolidated reverse modelling processes based on scan-to-BIM solutions. The most relevant contributions of the research reside in the study of the chances of using and adapting the characteristics of the IFC (Industry Foundation Classes) standard to the archiving needs of the diagnostic investigations results so that the digital twin model can meet the requirements of replicability in the context of the architectural heritage and interoperability with respect to the subsequent intervention phases envisaged by the conservation plan. Another crucial innovation is a proposal of a scan-to-BIM process improved by an automated approach performed by VPL (Visual Programming Languages) contribution. Finally, an online visualisation tool enables the HBIM cognitive system to be accessible and shareable by stakeholders involved in the general conservation process. [ABSTRACT FROM AUTHOR]

Due to the increase in aging structures and the decrease in construction workforce, there is an increasing interest in automating structural damage monitoring. Surface damage on concrete structures, such as cracks, delamination, and rebar exposure, is one of the important parameters that can be used to estimate the condition of the structure. In this paper, deep learning-based detection and quantification of structural damage using structured lights and a depth camera is proposed. The proposed monitoring system is composed of four lasers and a depth camera. The lasers are projected on the surface of the structures, and the camera captures images of the structures while measuring distance. By calculating an image homography, the captured images are calibrated when the structure and sensing system are not in parallel. The Faster RCNN (Region-based Convolutional Neural Network) with Inception Resnet v2 architecture is used to detect three types of surface damage: (i) cracks; (ii) delamination; and (iii) rebar exposure. The detected damage is quantified by calculating the positions of the projected laser beams with the measured distance. The experimental results show that structural damage was detected with an F1 score of 0.83 and a median value of the quantified relative error of less than 5%. [ABSTRACT FROM AUTHOR]

This paper proposes a customized convolutional neural network for crack detection in concrete structures. The proposed method is compared to four existing deep learning methods based on training data size, data heterogeneity, network complexity, and the number of epochs. The performance of the proposed convolutional neural network (CNN) model is evaluated and compared to pretrained networks, i.e., the VGG-16, VGG-19, ResNet-50, and Inception V3 models, on eight datasets of different sizes, created from two public datasets. For each model, the evaluation considered computational time, crack localization results, and classification measures, e.g., accuracy, precision, recall, and F1-score. Experimental results demonstrated that training data size and heterogeneity among data samples significantly affect model performance. All models demonstrated promising performance on a limited number of diverse training data; however, increasing the training data size and reducing diversity reduced generalization performance, and led to overfitting. The proposed customized CNN and VGG-16 models outperformed the other methods in terms of classification, localization, and computational time on a small amount of data, and the results indicate that these two models demonstrate superior crack detection and localization for concrete structures. [ABSTRACT FROM AUTHOR]

Three-point bending experiments of concrete beams were conducted under the strain rate range of 10−6 s−1 and 1.5 × 10−3 s−1. A novel 3D laser scanner, Handy SCAN, was employed to detect the areas of interface, mortar and aggregate on the crack surface after the experiment. In this paper, the inhomogeneity of materials and the inertial effect were considered as the main factors in the strength enhancement of concrete together with a proposed dynamic model. With the obtained experimental results, the initial elastic modulus and tensile strength of concrete showed obvious rate sensitivity. Moreover, an empirical relationship of dynamic increase factor and strain rate was established for the strain rate range of 10−6 s−1 and 1.5 × 10−3 s−1. The contributions of aggregate and inertia effect to the dynamic enhancement of concrete strength were quantified with respect to the loading rate. The rate effect of concrete obtained by the experiments was verified by the finite element analysis on the mesoscopic scale with the model built by the three-dimensional random aggregate software. [ABSTRACT FROM AUTHOR]

In order to achieve effective monitoring of concrete structures for sound structural health, the addition of carbon nanotubes (CNTs) into cementitious materials offers a promising solution for fabricating CNT-modified smart concrete with self-sensing ability. This study investigated the influences of CNT dispersion method, water/cement (W/C) ratio, and concrete constituents on the piezoelectric properties of CNT-modified cementitious materials. Three CNT dispersion methods (direct mixing, sodium dodecyl benzenesulfonate (NaDDBS) and carboxymethyl cellulose (CMC) surface treatment), three W/C ratios (0.4, 0.5, and 0.6), and three concrete constituent compositions (pure cement, cement/sand, and cement/sand/coarse aggregate) were considered. The experimental results showed that CNT-modified cementitious materials with CMC surface treatment had valid and consistent piezoelectric responses to external loading. The piezoelectric sensitivity improved significantly with increased W/C ratio and reduced progressively with the addition of sand and coarse aggregates. [ABSTRACT FROM AUTHOR]

Fibre optic sensors offer a means for the real-time continuous measurement of temperature or strain in concrete structures. Backscattered light along a fibre optic sensing (FOS) cable is interrogated to record a frequency shift and this shift is typically translated into a physical parameter such as strain or temperature using a calibration factor. However, when the measured frequency shift is a response to a combination of mechanical, thermal or hygral (humidity) loadings it is difficult to decouple individual influences. This presents a challenge in complex materials such as concrete where the strain, temperature and moisture levels change concurrently during the fresh and hardened states. Furthermore, depending on the application, both short- and longer-term measurements are required. As such, not only is the influence of these physical factors of interest but also the time and spatial stability of the measured frequency, which is highly dependent on the FOS cable composition. To investigate this aspect, fibre optic cables commonly used for strain (three tight-buffered cables) or temperature (two loose-buffered cables) measurement were considered. The cables were subjected to mechanical or environmental exposure and interrogated using a high-resolution optical backscatter reflectometer. The exposure regimes included three temperature cycles with sustained steps from 10 °C to 60 °C and back to 10 °C and an increasing and decreasing humidity cycle with steps between 30 to 90% relH. These ranges were selected to be indicative of typical environments for concrete. The results showed that the calibration factors back-calculated from increasing and decreasing temperature or humidity cycles differed. The third temperature cycle results were found to exhibit the smallest differences between heating and cooling suggesting that temperature pre-conditioning prior to installation could be advantageous. For all the cables, a drift in the readings was observed over the duration (2.5 h for temperature and 30 h for moisture) of the sustained steps. The magnitude of the drift depended on the cable type and exposure condition. In addition, local frequency fluctuations along the cable were observed which would need to be taken into account if only a single point along the cable length was used for analysis. The obtained results highlight the importance of the cable selection to maximise the FOS measurement fidelity for a given parameter of interest. [ABSTRACT FROM AUTHOR]

A vast amount of civil infrastructure is constructed using reinforced concrete, which can be susceptible to corrosion, posing significant risks. Corrosion of reinforced concrete has various causes, with chloride ingress known to be a major contributor. Monitoring this chloride ingress would allow for preventative maintenance to be less intrusive at a lower cost. Currently, chloride sensing methods are bulky and expensive, leaving the majority of concrete infrastructures unmonitored. This paper presents the design and fabrication of a miniature, low-cost device that can be embedded into concrete at various locations and depths. The device measures localized concrete resistance, correlating to the chloride ingress in the concrete using equations listed in this paper, and calculated results from two experiments are presented. The device benefits from a four-probe architecture, injecting a fixed frequency AC waveform across its outer electrodes within the cement block. Voltage across the internal electrodes is measured with a microcontroller and converted to a resistance value, communicated serially to an external computer. A final test showcases the ability of the device for three-dimensional mass deployment. [ABSTRACT FROM AUTHOR]

The goal of the condition assessment of concrete structures is to gain an insight into current condition of concrete and the existence of defects, which decrease durability and usability of the structure. Defects are quite difficult to detect using infrared thermography when concrete elements cannot be thermally excited with the Sun, together with unfavorable thermophysical properties of concrete structures. In this paper, principal component thermography (PCT) is applied as a post-processing method to a sequence of thermograms in order to enhance defect detectability in concrete structures. Defects are detected by analyzing a set of empirical orthogonal functions (EOFs), which were acquired by applying principal component analysis to a sequence of thermograms. The research was performed using concrete samples containing known defects, which were tested using a step heating thermography setup. The results of presented research show that PCT is an effective post-processing method to improve defect detection in concrete structures. By effectively improving the defect detection, PCT has a potential to improve the non-destructive testing (NDT) accuracy of using infrared thermography (IRT) on concrete structures, especially in shaded areas of such structures. The research also shows the defect detectability depending on concrete type thermal excitation setup and defect geometry. [ABSTRACT FROM AUTHOR]

This paper proposes a new approach to damage detection of nano-SiO2 concrete-filled glass fiber reinforced polymer (GFRP) tube column using piezoceramic transducers. Stress waves are emitted and received by a pair of piezoceramic transducers embedded in the concrete-filled GFRP tube, and the energy and damage indices at different levels of loading in the tube are obtained by wavelet packet to evaluate the damage degree of GFRP tube nano-SiO2 concrete column. Through the experimental studies, the effects of different nano-SiO2 contents, concrete grades, and superplasticizer on the damage were analyzed to gain load–displacement curves, load–energy index curves, and load–damage index curves. The results show that the wave method can be adopted to monitor the damage of GFRP tube nano-SiO2 concrete column. The specimens with 3% nano-SiO2 content have the smallest energy change rate, indicating that adding 3% nano-SiO2 content into concrete can effectively delay the development of damage. After the addition of superplasticizer, with the increase in the strength grade of concrete, the cracks in the specimen tend to develop slowly, and therefore the specimens have a stronger resistance to damage. The damage of the specimens with the nano-SiO2 content of 1% appeared the latest, while the damage without the nano-SiO2 specimen appeared the fastest. The experimental results show that this method can better monitor the damage of the Nano-SiO2 concrete in the glass fiber reinforced polymer (GFRP) tube. [ABSTRACT FROM AUTHOR]

In this paper, we study the strain transfer mechanism between a host material and an optical fiber. A new analytical model handling imperfect bonding between layers is proposed. A general expression of the crack-induced strain transfer from fractured concrete material to optical fiber is established in the case of a multilayer system. This new strain transfer model is examined through performing wedge splitting tests on concrete specimens instrumented with embedded and surface-mounted fiber optic cables. The experimental results showed the validity of the crack-induced strain expression fitted to the distributed strains measured using an Optical Backscattering Reflectometry (OBR) system. As a result, precise estimations of the crack openings next to the optical cable location were achieved, as well as the monitoring of the optical cable response through following the strain lag parameter. [ABSTRACT FROM AUTHOR]

Slip-forming in concrete construction enables the continuous placement of concrete using a climbing form, the efficiency of which depends on appropriate slip-up timing. This implies the importance of knowing accurately the development of concrete strength over time, which has been assessed manually to date in construction fields. This paper presents a method for automating the slip-forming process by determining the optimal slip-up time using the in-situ strength of concrete. The strength of concrete is evaluated by a formula relating the strength to the surface wave velocity measured with ultrasonic sensors. Specifically, this study validates the applicability of the slip-form system with ultrasonic sensors for continuously monitoring the hardening of concrete through its application in several construction sites. To this end, a slip-form system with a pair of ultrasonic modules at the bottom of the panel was tested and the time variation of surface wave velocity in the concrete material was monitored during the slip-forming process. The results show that the proposed method can provide the optimal slip-up time of the form to automate the slip-forming process. This approach is expected to apply to other construction technologies that required the continuous monitoring of concrete strength for construction efficiency as well as quality maintenance. [ABSTRACT FROM AUTHOR]

In this paper, a magnetic microwire-based sensor array embedded under the pavement is proposed as a weighing system at customs ports of entry. This sensor is made of a cementitious material suitable for embedding within the core of concrete structures prior to curing. The objective of this research is to verify the feasibility of stress monitoring for concrete materials using an array of cement-based stress/strain sensors that have been developed using the magnetic sensing property of an embedded microwire in a cement-based composite. Test results for microwire-based sensors and gauge sensors are compared. The strain sensitivity and their linearity are investigated through experimental testing under compressive loadings. Sensors made of these materials can be designed to satisfy specific needs and reduce costs in the production of sensor aggregates with improved coupling performance, thus avoiding any disturbance to the stress state. [ABSTRACT FROM AUTHOR]

The recent application of deep learning for structural health monitoring systems for damage detection has potential for improvised structure performance and maintenance for long term durability, and reliable strength. Advancements in electro-mechanical impedance (EMI) techniques have sparked attention among researchers to develop novel monitoring techniques for structural monitoring and evaluation. This study aims to determine the performance of EMI techniques using a piezo sensor to monitor the development of bond strength in reinforced concrete through a pull-out test. The concrete cylindrical samples with embedded steel bars were prepared, cured for 28 days, and a pull-out test was performed to measure the interfacial bond between them. The piezo coupled signatures were obtained for the PZT patch bonded to the steel bar. The damage qualification is performed through the statistical indices, i.e., root-mean-square deviation (RMSD) and correlation coefficient deviation metric (CCDM), were obtained for different displacements recorded for axial pull. Furthermore, this study utilizes a novel Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM)-based hybrid model, an effective regression model to predict the EMI signatures. These results emphasize the efficiency and potential application of the deep learning-based hybrid model in predicting EMI-based structural signatures. The findings of this study have several implications for structural health diagnosis using a deep learning-based model for monitoring and conservation of building heritage. [ABSTRACT FROM AUTHOR]

Shrinkage is an important component of the behaviour of reinforced concrete (RC) structures, however, the number of variables that affect shrinkage make it a complex time-dependent phenomenon. Additionally, as new concrete materials with lower embodied carbon gain popularity, there is a need for an in-depth understanding into their shrinkage behaviour before they can be widely adopted by industry. Currently, the shrinkage behaviour of concrete is studied using discrete measurements on small-scale unrestrained prisms. Distributed fibre optic sensing (DFOS) potentially provides a method of measuring both restrained (with reinforcement) and unrestrained (without reinforcement) shrinkage in both small-scale specimens and structural elements. In the current study, methods of measuring distributed unrestrained shrinkage strains were developed and evaluated, and the restrained shrinkage strains in different types of structural members were studied. Unrestrained shrinkage strains were measured using fibres optic cables embedded in small concrete prisms, while restrained shrinkage strains were measured with fibres bonded to the longitudinal reinforcement. Unrestrained shrinkage strains were found to be highly variable (as large as 3800 microstrain range) depending on location, but further research needs to be undertaken to account for end effects, early-stage shrinkage, and bond between the fibre optic cable and the concrete. Restrained shrinkage strains from structural members revealed non-uniform shrinkage strain distributions along member length due to functional grading as well as high supplementary cementitious material concretes, suggesting that shrinkage models will need to account for this variability. [ABSTRACT FROM AUTHOR]

In order to realize the automatic classification of internal defects for non-contact nondestructive testing of concrete, a concrete multi-type defect classification algorithm based on the mixed strategy slime mold algorithm support vector machine (MSSMA-SVM) was proposed. The concrete surface's vibration signal was obtained using a laser Doppler vibrometer (LDV) for four classification targets for no defect, segregation, cavity, and foreign matter concrete classification targets. The wavelet packet transform (WPT) decomposes the detected signals to get information on different frequency bands. The energy ratio change rate, energy ratio, and wavelet packet singular entropy of each node after the WPT were used as the feature input of MSSMA-SVM. The experimental results show that the designed MSSMA-SVM classifier can accurately detect the type, which provides a practical algorithm for classifying concrete defects by laser vibration measurement. [ABSTRACT FROM AUTHOR]

The article presents research on the performance of composite and monolithic sensors for distributed fibre optic sensing (DFOS). The introduction summarises the design of the sensors and the theoretical justification for such an approach. Lessons learned during monitoring cracked concrete are summarised to highlight what features of the DFOS tools are the most favourable from the crack analysis point of view. Later, the results from full-size laboratory concrete specimens working in a cracked state were presented and discussed in reference to conventional layered sensing cables. The research aimed to compare monolithic sensors and layered cables embedded in the same reinforced concrete elements, which is the main novelty. The performance of each DFOS nondestructive tool was investigated in the close vicinity of the cracks—both the new ones, opening within the tension zone, and the existing ones, closing within the compression zone. The qualitative (detection) and quantitative (widths estimation) crack analyses were performed and discussed. Finally, the examples of actual applications within concrete structures, including bridges, are presented with some examples of in situ results. [ABSTRACT FROM AUTHOR]

This paper proposes a static stress estimation method for concrete structures, using the stress relaxation method (SRM) in conjunction with digital image correlation (DIC). The proposed method initially requires a small hole to be drilled in the concrete surface to induce stress relaxation around the hole and, consequently, a displacement field. DIC measures this displacement field by comparing digital images taken before and after the hole-drilling. The stress level in the concrete structure is then determined by solving an optimization problem, designed to minimize the difference between the displacement fields from DIC and the one from a numerical model. Compared to the pointwise measurements by strain gauges, the full-field displacement obtained by DIC provides a larger amount of data, leading to a more accurate estimation. Our theoretical results were experimentally validated using concrete specimens, demonstrating the efficacy of the proposed method. [ABSTRACT FROM AUTHOR]

Reinforced concrete structures' (RCSs) ageing and early deterioration are some of the main challenges faced by the building sector today, and steel bar corrosion is one of the main problems. In this phenomenon, water and concrete's electric resistivity play a fundamental role. Therefore, developing sensor systems capable of estimating any variations in these parameters in real time and remotely would represent considerable progress in sustainably maintaining RCSs. Many types of sensors capable of estimating humidity variation and electrical resistivity in concrete currently exist, but the variability of these sensors' sensitivity can be extreme depending on several factors; for example, temperature or presence of ions and their incorporation into smart monitoring systems, which is difficult. As an alternative to today's sensors, this study centered on developing two estimation models by means of the response of a novel voltammetric stainless steel (SS) sensor. The estimation models were one of humidity variation and another of concrete's electric resistivity. These models were calibrated, fitted and validated. In the validation, both these models explained a percentage of variance over 80%. [ABSTRACT FROM AUTHOR]

Concrete exhibits time-dependent long-term behavior driven by creep and shrinkage. These rheological effects are difficult to predict due to their stochastic nature and dependence on loading history. Existing empirical models used to predict rheological effects are fitted to databases composed largely of laboratory tests of limited time span and that do not capture differential rheological effects. A numerical model is typically required for application of empirical constitutive models to real structures. Notwithstanding this, the optimal parameters for the laboratory databases are not necessarily ideal for a specific structure. Data-driven approaches using structural health monitoring data have shown promise towards accurate prediction of long-term time-dependent behavior in concrete structures, but current approaches require different model parameters for each sensor and do not leverage geometry and loading. In this work, a physics-informed data-driven approach for long-term prediction of 2D normal strain field in prestressed concrete structures is introduced. The method employs a simplified analytical model of the structure, a data-driven model for prediction of the temperature field, and embedding of neural networks into rheological time-functions. In contrast to previous approaches, the model is trained on multiple sensors at once and enables the estimation of the strain evolution at any point of interest in the longitudinal section of the structure, capturing differential rheological effects. [ABSTRACT FROM AUTHOR]

A recently developed contactless ultrasonic testing scheme is applied to define the optimal saw-cutting time for concrete pavement. The ultrasonic system is improved using wireless data transfer for field applications, and the signal processing and data analysis are proposed to evaluate the modulus of elasticity of early-age concrete. Numerical simulation of leaky Rayleigh wave in joint-half space including air and concrete is performed to demonstrate the proposed data analysis procedure. The hardware and algorithms developed for the ultrasonic system are experimentally validated with a comparison of saw-cutting procedures. In addition, conventional methods for the characterization of early-age concrete, including pin penetration and maturity methods, are applied. The results demonstrated that the developed wireless system presents identical results to the wired system, and the initiation time of leaky Rayleigh wave possibly represents 5% of raveling damage compared to the optimal saw cutting. Further data analysis implies that saw-cutting would be optimally performed at approximately 11.5 GPa elastic modulus of concrete obtained by the wireless and contactless ultrasonic system. [ABSTRACT FROM AUTHOR]

Concrete constructions need widespread monitoring for the control of their state of integrity during their service life. In particular, after critical events such as earthquakes, this type of structure may experience the formation and development of cracks and damage. A quick and affordable assessment of structural behavior is indicated to identify conditions of danger for users and the incipient collapse of structural elements. This work presents investigations on multifunctional concretes with self-sensing capabilities to carry out static and dynamic monitoring. The materials were produced by the addition of conductive carbon microfibers to the concrete matrix. Electrical and sensing tests were carried out on samples with small-, medium-, and full-scale dimensions. The tests demonstrated the good electrical and electromechanical properties of the proposed smart concrete sensors, which appear promising for their use in civil elements or structures. In particular, tests on real-scale beams demonstrated the capability of the material to monitor the dynamic behavior of full-scale structural elements. [ABSTRACT FROM AUTHOR]

The Impact-Echo (IE) test is an effective method for determining the presence, depth, and area of cracks in concrete as well as the dimensions of the sound concrete without defects. In addition, shallow delamination can be measured by confirming a flexural mode in the low-frequency region. Owing to the advancement of non-contact sensors and automated measurement equipment, the IE test can be measured at multiple points in a short period. To analyze and distinguish a large volume of data, applying supervised learning (SL) associated with various contemporary algorithms is necessary. However, SL has limitations due to the difficulty in accurate labeling for increased volumes of test data, and reflection of new specimen characteristics, and it is necessary to apply semi-supervised learning (SSL) to overcome them. This study analyzes the accuracy and evaluates the applicability of a model trained with SSL rather than SL using the data from the air-coupled IE test based on dynamic preconditions. For the detection of delamination defects, the dynamic behavior-based flexural mode was identified, and 21 features were extracted in the time and frequency domains. Three principal components (PCs) such as the real moment, real RMS, and imaginary moment were derived through principal component analysis (PCA). PCs were identical in slab, pavement, and deck. In the case of SSL considering a dynamic behavior, the accuracy increased by 7–8% compared with SL, and it could categorize good, fair, and poor status to a higher level for actual structures. The applicability of SSL to the IE test was confirmed, and because the crack progress varies under field conditions, other parameters must be considered in the future to reflect this. [ABSTRACT FROM AUTHOR]

The roughness of a surface is a decisive parameter of a material. In rehabilitation of concrete structures, for example, it significantly affects the adhesion between the coating material and the base concrete. However, the standard measurement procedure in construction suffers from considerable disadvantages, which leads to the demand for more sophisticated methods. In a research project, we, therefore, developed a novel camera-based measurement system, which is customized to meet the prevailing requirements for practical use on construction sites. In this article, we provide an overview of the measurement system and present comprehensive examinations to evaluate the accuracy and to provide evidence of validity. First, we examined the accuracy of the system by empirically assessing both trueness and precision of measurements using three concrete specimens. Trueness was determined by comparing the surface measurements to those of a highly accurate microscope system, revealing RMSE values of around 40–50 µm. Precision, on the other hand, was assessed considering the scattering of the roughness measurements under repeat conditions, which led to standard deviations of less than 6 µm. Furthermore, to proof validity, a comparative study was conducted based on sixteen concrete specimens, which includes the sand patch method and laser triangulation as established roughness measurement methods in practice. The empirically determined correlation coefficients between all three methods were greater than 0.99, indicating extraordinarily high linear relationships. Among them, the greatest correlation was between the camera-based system and laser triangulation. [ABSTRACT FROM AUTHOR]

Semantic segmentation network-based methods can detect concrete damage at the pixel level. However, the performance of a single semantic segmentation network is often limited. To improve the concrete damage detection performance of a semantic segmentation network, a stacking ensemble learning-based concrete crack detection method using multiple semantic segmentation networks is proposed. To realize this method, a database including 500 images and their labels with concrete crack and spalling is built and divided into training and testing sets. At first, the training and prediction of five semantic segmentation networks (FCN-8s, SegNet, U-Net, PSPNet and DeepLabv3+) are respectively implemented on the built training set according to a five-fold cross-validation principle, where 80% of the training images are used in the training process, and 20% training images are reserved. Then, in predicting the results of reserved training images from trained semantic segmentation networks, the class labels of all pixels are collected, and then four softmax regression-based ensemble learning models are trained using the collected class labels and their true classification labels. The trained ensemble learning models are applied to regressed testing results of semantic segmentation network models. Compared with the best single semantic segmentation network, the best ensemble learning model provides performance improvement of 0.21% PA, 0.54% MPA, 3.66% MIoU, and 0.12% FWIoU, respectively. The study results show that the stacking ensemble learning strategy can indeed improve concrete damage detection performance through ensemble learning of multiple semantic segmentation networks. [ABSTRACT FROM AUTHOR]