Estimation of Prediction Intervals for Performance Assessment of Building Using Machine Learning.

This study utilizes artificial neural networks (ANN) to estimate prediction intervals (PI) for seismic performance assessment of buildings subjected to long-term ground motion. To address the uncertainty quantification in structural health monitoring (SHM), the quality-driven lower upper bound estimation (QD-LUBE) has been opted for global probabilistic assessment of damage at local and global levels, unlike traditional methods. A distribution-free machine learning model has been adopted for enhanced reliability in quantifying uncertainty and ensuring robustness in post-earthquake probabilistic assessments and early warning systems. The distribution-free machine learning model is capable of quantifying uncertainty with high accuracy as compared to previous methods such as the bootstrap method, etc. This research demonstrates the efficacy of the QD-LUBE method in complex seismic risk assessment scenarios, thereby contributing significant enhancement in building resilience and disaster management strategies. This study also validates the findings through fragility curve analysis, offering comprehensive insights into structural damage assessment and mitigation strategies.

Fast Detection of Missing Thin Propagating Cracks during Deep-Learning-Based Concrete Crack/Non-Crack Classification.

Existing deep learning (DL) models can detect wider or thicker segments of cracks that occupy multiple pixels in the width direction, but fail to distinguish the thin tail shallow segment or propagating crack occupying fewer pixels. Therefore, in this study, we proposed a scheme for tracking missing thin/propagating crack segments during DL-based crack identification on concrete surfaces in a computationally efficient manner. The proposed scheme employs image processing as a preprocessor and a postprocessor for a 1D DL model. Image-processing-assisted DL as a precursor to DL eliminates labor-intensive labeling and the plane structural background without any distinguishable features during DL training and testing; the model identifies potential crack candidate regions. Iterative differential sliding-window-based local image processing as a postprocessor to DL tracks missing thin cracks on segments classified as cracks. The capability of the proposed method is demonstrated on low-resolution images with cracks of single-pixel width, captured using unmanned aerial vehicles on concrete structures with different surface textures, different scenes with complicated disturbances, and optical variability. Due to the multi-threshold-based image processing, the overall approach is invariant to the choice of initial sensitivity parameters, hyperparameters, and the sequence of neuron arrangement. Further, this technique is a computationally efficient alternative to semantic segmentation that results in pixelated mapping/classification of thin crack regimes, which requires labor-intensive and skilled labeling.

Damage-Detection Approach for Bridges with Multi-Vehicle Loads Using Convolutional Autoencoder.

Deep learning has been widely employed in recent studies on bridge-damage detection to improve the performance of damage-detection methods. Unsupervised deep learning can be effectively utilized to increase the applicability of damage-detection approaches. Hence, the authors propose a convolutional-autoencoder (CAE)-based damage-detection approach, which is an unsupervised deep-learning network. However, the CAE-based damage-detection approach demonstrates only satisfactory accuracy for prestressed concrete bridges with a single-vehicle load. Therefore, this study was performed to verify whether the CAE-based damage-detection approach can be applied to bridges with multi-vehicle loads, which is a typical scenario. In this study, rigid-frame and reinforced-concrete-slab bridges were modeled and simulated to obtain the behavior data of bridges. A CAE-based damage-detection approach was tested on both bridges. For both bridges, the results demonstrated satisfactory damage-detection accuracy of over 90% and a false-negative rate of less than 1%. These results prove that the CAE-based approach can be successfully applied to various types of bridges with multi-vehicle loads.

Special Issue on "Smart City and Smart Infrastructure".

Recent developments in sensor technologies and data-driven approaches have been recognized as the main enablers of smart cities [...].

LiDAR-Based Bridge Displacement Estimation Using 3D Spatial Optimization.

As civil engineering structures become larger, non-contact inspection technology is required to measure the overall shape and size of structures and evaluate safety. Structures are easily exposed to the external environment and may not be able to perform their original functions depending on the continuous load for a long time. Therefore, in this study, we propose a method for estimating the vertical displacement of structures using light detection and ranging, which enables non-contact measurement. The point cloud acquired through laser scanning was rearranged into a three-dimensional space, and internal nodes were created by continuously dividing the space. The generated node has its own location information, and the vertical displacement value was calculated by searching for the node where the deformation occurred. The performance of the proposed displacement estimation technique was verified through static loading experiments, and the octree space partitioning method is expected to be applied and utilized in structural health monitoring.

Individual Disaster Assistance For Socially Vulnerable People: Lessons Learned From the Pohang Earthquake in the Republic of Korea.

The Republic of Korea has been considered to be relatively safe from earthquake hazards because of the geological location of the Korean Peninsula, which has a low level of intraplate seismic activity. However, an earthquake with a moment magnitude of 5.4 struck the city of Pohang on November 15, 2017, causing 90 casualties and 52 million USD in property losses. During the recovery process after the earthquake, the Korean government provided individual disaster assistance to victims who reported their damages. However, the government disaster assistance could have been unfairly distributed among the socially vulnerable victims who essentially relied on that assistance. This study identifies whether the government disaster assistance was fairly distributed to socially vulnerable victims using a statistical model based on the data from the Pohang earthquake that occurred in 2017 in Korea. A conceptual model was constructed using a structural equation model (SEM) of three factors-social vulnerability, physical vulnerability, and the amount paid out in individual disaster assistance. Furthermore, interviews with and a survey of the victims were conducted to verify the problems identified by the conceptual model. This study found that socially vulnerable victims were less likely to take advantage of the government disaster assistance program.

Prediction of Static Modulus and Compressive Strength of Concrete from Dynamic Modulus Associated with Wave Velocity and Resonance Frequency Using Machine Learning Techniques.

The static elastic modulus (Ec) and compressive strength (fc) are critical properties of concrete. When determining Ec and fc, concrete cores are collected and subjected to destructive tests. However, destructive tests require certain test permissions and large sample sizes. Hence, it is preferable to predict Ec using the dynamic elastic modulus (Ed), through nondestructive evaluations. A resonance frequency test performed according to ASTM C215-14 and a pressure wave (P-wave) measurement conducted according to ASTM C597M-16 are typically used to determine Ed. Recently, developments in transducers have enabled the measurement of a shear wave (S-wave) velocities in concrete. Although various equations have been proposed for estimating Ec and fc from Ed, their results deviate from experimental values. Thus, it is necessary to obtain a reliable Ed value for accurately predicting Ec and fc. In this study, Ed values were experimentally obtained from P-wave and S-wave velocities in the longitudinal and transverse modes; Ec and fc values were predicted using these Ed values through four machine learning (ML) methods: support vector machine, artificial neural networks, ensembles, and linear regression. Using ML, the prediction accuracy of Ec and fc was improved by 2.5-5% and 7-9%, respectively, compared with the accuracy obtained using classical or normal-regression equations. By combining ML methods, the accuracy of the predicted Ec and fc was improved by 0.5% and 1.5%, respectively, compared with the optimal single variable results.

Characterization of Porous Cementitious Materials Using Microscopic Image Processing and X-ray CT Analysis.

The use of lightweight concrete has continuously increased because it has a primary benefit of reducing dead load in a concrete infrastructure. Various properties of lightweight concrete, such as compressive strength, elastic modulus, sound absorption performance, and thermal insulation, are highly related to its pore characteristics. Consequently, the identification of the characteristics of its pores is an important task. This study performs a comparative analysis for characterizing the pores in cementitious materials using three different testing methods: a water absorption test, microscopic image processing, and X-ray computed tomography (X-ray CT) analysis. For all 12 porous cementitious materials, conventional water absorption test was conducted to obtain their water permeable porosities. Using the microscopic image processing method, various characteristics of pores were identified in terms of the 2D pore ratio (i.e., ratio of pore area to total surface area), the pore size, and the number of pores in the cross-sectional area. The 3D tomographic image-based X-ray CT analysis was conducted for the selected samples to show the 3D pore ratio (i.e., ratio of pore volume to total volume), the pore size, the spatial distribution of pores along the height direction of specimen, and open and closed pores. Based on the experimental results, the relationships of oven-dried density with these porosities were identified. Research findings revealed that the complementary use of these testing methods is beneficial for analyzing the characteristics of pores in cementitious materials.

Stress Estimation Using Digital Image Correlation with Compensation of Camera Motion-Induced Error.

Measurement of stress levels from an in-service structure can provide important and useful information regarding the current state of a structure. The stress relaxation method (SRM) is the most conventional and practical method, which has been widely accepted for measuring residual stresses in metallic materials. The SRM showed strong potential for stress estimation of civil engineering structures, when combined with digital image correlation (DIC). However, the SRM/DIC methods studied thus far have practical issues regarding camera vibration during hole drilling. To minimize the error induced by the camera motion, the imaging system is installed at a distance from the specimen resulting in the low pixel density and the large extent of the inflicted damage. This study proposes an SRM/DIC-based stress estimation method that allows the camera to be removed during hole drilling and relocated to take the after-drilling image. Since the imaging system can be placed as close to the specimen as possible, a high pixel density can be achieved such that subtle displacement perturbation introduced by a small damage can be acquired by DIC. This study provides a detailed mathematical formulation for removing the camera relocation-induced false displacement field in the DIC result. The proposed method is validated numerically and experimentally.

Bayesian Prediction of Pre-Stressed Concrete Bridge Deflection Using Finite Element Analysis.

Vertical deflection has been emphasized as an important safety indicator in the management of railway bridges. Therefore, various standards and studies have suggested physics-based models for predicting the time-dependent deflection of railway bridges. However, these approaches may be limited by model errors caused by uncertainties in various factors, such as material properties, creep coefficient, and temperature. This study proposes a new Bayesian method that employs both a finite element model and actual measurement data. To overcome the limitations of an imperfect finite element model and a shortage of data, Gaussian process regression is introduced and modified to consider both, the finite element analysis results and actual measurement data. In addition, the probabilistic prediction model can be updated whenever additional measurement data is available. In this manner, a probabilistic prediction model, that is customized to a target bridge, can be obtained. The proposed method is applied to a pre-stressed concrete railway bridge in the construction stage in the Republic of Korea, as an example of a bridge for which accurate time-dependent deflection is difficult to predict, and measurement data are insufficient. Probabilistic prediction models are successfully derived by applying the proposed method, and the corresponding prediction results agree with the actual measurements, even though the bridge experienced large downward deflections during the construction stage. In addition, the practical uses of the prediction models are discussed.

Prediction Model for Mechanical Properties of Lightweight Aggregate Concrete Using Artificial Neural Network.

The mechanical properties of lightweight aggregate concrete (LWAC) depend on the mixing ratio of its binders, normal weight aggregate (NWA), and lightweight aggregate (LWA). To characterize the relation between various concrete components and the mechanical characteristics of LWAC, extensive studies have been conducted, proposing empirical equations using regression models based on their experimental results. However, these results obtained from laboratory experiments do not provide consistent prediction accuracy due to the complicated relation between materials and mix proportions, and a general prediction model is needed, considering several mix proportions and concrete constituents. This study adopts the artificial neural network (ANN) for modeling the complex and nonlinear relation between constituents and the resulting compressive strength and elastic modulus of LWAC. To construct a database for the ANN model, a vast amount of detailed and extensive data was collected from the literature including various mix proportions, material properties, and mechanical characteristics of concrete. The optimal ANN architecture is determined to enhance prediction accuracy in terms of the numbers of hidden layers and neurons. Using this database and the optimal ANN model, the performance of the ANN-based prediction model is evaluated in terms of the compressive strength and elastic modulus of LWAC. Furthermore, these prediction accuracies are compared to the results of previous ANN-based analyses, as well as those obtained from the commonly used linear and nonlinear regression models.

A Novelty Detection Approach for Tendons of Prestressed Concrete Bridges Based on a Convolutional Autoencoder and Acceleration Data.

The most important structural element of prestressed concrete (PSC) bridges is the prestressed tendon, and in order to ensure safety of such bridges, it is very important to determine whether the tendon is damaged. However, it is not easy to detect tendon damage in real time. This study proposes a novelty detection approach for damage to the tendons of PSC bridges based on a convolutional autoencoder (CAE). The proposed method employs simulation data from nine accelerometers. The accuracies of CAEs for multi-vehicle are 79.5%⁻85.8% for 100% and 75% damage severities with all error levels and 50% damage severity without error. However, the accuracies for 50% damage severity with 5% and 10% error levels drop to 69.4%⁻73.3%. The accuracies of CAEs for single-vehicle ranges from 90.1%⁻95.1% for all damage severities and error levels that are satisfactory. The findings indicate that the CAE approach for multi-vehicle can be effective when the damages are severe, but not when moderate. Meanwhile, if acceleration data can be obtained for single-vehicle, then the CAE approach can provide a highly accurate and robust method of tendon damage detection in PSC bridges in use, even if the measurement errors are significant.

Uniaxial Static Stress Estimation for Concrete Structures Using Digital Image Correlation.

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.

Applicability of Diffuse Ultrasound to Evaluation of the Water Permeability and Chloride Ion Penetrability of Cracked Concrete.

This study aims to explore the applicability of diffuse ultrasound to the evaluation of water permeability and chloride ion penetrability of cracked concrete. Lab-scale experiments were conducted on disk-shaped concrete specimens, each having a different width of a penetrating crack that was generated by splitting tension along the centerline. The average crack width of each specimen was determined using an image binarization technique. The diffuse ultrasound test employed signals in the frequency range of 200 to 440 kHz. The water flow rate was measured using a constant water-head permeability method, and the chloride diffusion coefficient was determined using a modified steady-state migration method. Then, the effects of crack width on the diffusion characteristics of ultrasound (i.e., diffusivity, dissipation), water flow rate, and chloride diffusion coefficient are investigated. The correlations between the water flow rate and diffuse ultrasound parameters, and between the chloride diffusion coefficient and diffuse ultrasound parameters, are examined. The results suggest that diffuse ultrasound is a promising method for assessing the water permeability and chloride ion penetrability of cracked concrete.

Surface-Wave Based Model for Estimation of Discontinuity Depth in Concrete.

In this paper, we propose an accurate and practical model for the estimation of surface-breaking discontinuity (i.e., crack) depth in concrete through quantitative characterization of surface-wave transmission across the discontinuity. The effects of three different mixture types (mortar, normal strength concrete, and high strength concrete) and four different simulated crack depths on surface-wave transmission were examined through experiments carried out on lab-scale concrete specimens. The crack depth estimation model is based on a surface-wave spectral energy approach that is capable of taking into account a wide range of wave frequencies. The accuracy of the proposed crack depth estimation model is validated by root mean square error analysis of data from repeated spectral energy transmission ratio measurements for each specimen.

Computer Vision-Based Structural Displacement Measurement Robust to Light-Induced Image Degradation for In-Service Bridges.

The displacement responses of a civil engineering structure can provide important information regarding structural behaviors that help in assessing safety and serviceability. A displacement measurement using conventional devices, such as the linear variable differential transformer (LVDT), is challenging owing to issues related to inconvenient sensor installation that often requires additional temporary structures. A promising alternative is offered by computer vision, which typically provides a low-cost and non-contact displacement measurement that converts the movement of an object, mostly an attached marker, in the captured images into structural displacement. However, there is limited research on addressing light-induced measurement error caused by the inevitable sunlight in field-testing conditions. This study presents a computer vision-based displacement measurement approach tailored to a field-testing environment with enhanced robustness to strong sunlight. An image-processing algorithm with an adaptive region-of-interest (ROI) is proposed to reliably determine a marker's location even when the marker is indistinct due to unfavorable light. The performance of the proposed system is experimentally validated in both laboratory-scale and field experiments.

Concrete Crack Identification Using a UAV Incorporating Hybrid Image Processing.

Crack assessment is an essential process in the maintenance of concrete structures. In general, concrete cracks are inspected by manual visual observation of the surface, which is intrinsically subjective as it depends on the experience of inspectors. Further, it is time-consuming, expensive, and often unsafe when inaccessible structural members are to be assessed. Unmanned aerial vehicle (UAV) technologies combined with digital image processing have recently been applied to crack assessment to overcome the drawbacks of manual visual inspection. However, identification of crack information in terms of width and length has not been fully explored in the UAV-based applications, because of the absence of distance measurement and tailored image processing. This paper presents a crack identification strategy that combines hybrid image processing with UAV technology. Equipped with a camera, an ultrasonic displacement sensor, and a WiFi module, the system provides the image of cracks and the associated working distance from a target structure on demand. The obtained information is subsequently processed by hybrid image binarization to estimate the crack width accurately while minimizing the loss of the crack length information. The proposed system has shown to successfully measure cracks thicker than 0.1 mm with the maximum length estimation error of 7.3%.

Principles and Applications of Ultrasonic-Based Nondestructive Methods for Self-Healing in Cementitious Materials.

Recently, self-healing technologies have emerged as a promising approach to extend the service life of social infrastructure in the field of concrete construction. However, current evaluations of the self-healing technologies developed for cementitious materials are mostly limited to lab-scale experiments to inspect changes in surface crack width (by optical microscopy) and permeability. Furthermore, there is a universal lack of unified test methods to assess the effectiveness of self-healing technologies. Particularly, with respect to the self-healing of concrete applied in actual construction, nondestructive test methods are required to avoid interrupting the use of the structures under evaluation. This paper presents a review of all existing research on the principles of ultrasonic test methods and case studies pertaining to self-healing concrete. The main objective of the study is to examine the applicability and limitation of various ultrasonic test methods in assessing the self-healing performance. Finally, future directions on the development of reliable assessment methods for self-healing cementitious materials are suggested.

Probabilistic Assessment of High-Throughput Wireless Sensor Networks.

Structural health monitoring (SHM) using wireless smart sensors (WSS) has the potential to provide rich information on the state of a structure. However, because of their distributed nature, maintaining highly robust and reliable networks can be challenging. Assessing WSS network communication quality before and after finalizing a deployment is critical to achieve a successful WSS network for SHM purposes. Early studies on WSS network reliability mostly used temporal signal indicators, composed of a smaller number of packets, to assess the network reliability. However, because the WSS networks for SHM purpose often require high data throughput, i.e., a larger number of packets are delivered within the communication, such an approach is not sufficient. Instead, in this study, a model that can assess, probabilistically, the long-term performance of the network is proposed. The proposed model is based on readily-available measured data sets that represent communication quality during high-throughput data transfer. Then, an empirical limit-state function is determined, which is further used to estimate the probability of network communication failure. Monte Carlo simulation is adopted in this paper and applied to a small and a full-bridge wireless networks. By performing the proposed analysis in complex sensor networks, an optimized sensor topology can be achieved.

Decentralized system identification using stochastic subspace identification for wireless sensor networks.

Wireless sensor networks (WSNs) facilitate a new paradigm to structural identification and monitoring for civil infrastructure. Conventional structural monitoring systems based on wired sensors and centralized data acquisition systems are costly for installation as well as maintenance. WSNs have emerged as a technology that can overcome such difficulties, making deployment of a dense array of sensors on large civil structures both feasible and economical. However, as opposed to wired sensor networks in which centralized data acquisition and processing is common practice, WSNs require decentralized computing algorithms to reduce data transmission due to the limitation associated with wireless communication. In this paper, the stochastic subspace identification (SSI) technique is selected for system identification, and SSI-based decentralized system identification (SDSI) is proposed to be implemented in a WSN composed of Imote2 wireless sensors that measure acceleration. The SDSI is tightly scheduled in the hierarchical WSN, and its performance is experimentally verified in a laboratory test using a 5-story shear building model.