• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.325-334
• /
• 2023

Computer vision-based damage detection enables non-contact, efficient and low-cost bridge health monitoring, which reduces the need for labor-intensive manual inspection or that for a large number of on-site sensing instruments. By leveraging recent semantic segmentation approaches, we can detect regions of critical structural components and identify damages at pixel level on images. However, existing methods perform poorly when detecting small and thin damages (e.g., cracks); the problem is exacerbated by imbalanced samples. To this end, we incorporate domain knowledge to introduce a hierarchical semantic segmentation framework that imposes a hierarchical semantic relationship between component categories and damage types. For instance, certain types of concrete cracks are only present on bridge columns, and therefore the noncolumn region may be masked out when detecting such damages. In this way, the damage detection model focuses on extracting features from relevant structural components and avoid those from irrelevant regions. We also utilize multi-scale augmentation to preserve contextual information of each image, without losing the ability to handle small and/or thin damages. In addition, our framework employs an importance sampling, where images with rare components are sampled more often, to address sample imbalance. We evaluated our framework on a public synthetic dataset that consists of 2,000 railway bridges. Our framework achieves a 0.836 mean intersection over union (IoU) for structural component segmentation and a 0.483 mean IoU for damage segmentation. Our results have in total 5% and 18% improvements for the structural component segmentation and damage segmentation tasks, respectively, compared to the best-performing baseline model.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.335-349
• /
• 2023

Bridges constantly undergo deterioration and damage, the most common ones being concrete damage and exposed rebar. Periodic inspection of bridges to identify damages can aid in their quick remediation. Likewise, identifying components can provide context for damage assessment and help gauge a bridge's state of interaction with its surroundings. Current inspection techniques rely on manual site visits, which can be time-consuming and costly. More recently, robotic inspection assisted by autonomous data analytics based on Computer Vision (CV) and Artificial Intelligence (AI) has been viewed as a suitable alternative to manual inspection because of its efficiency and accuracy. To aid research in this avenue, this study performs a comparative assessment of different architectures, loss functions, and ensembling strategies for the autonomous segmentation of bridge components and damages. The experiments lead to several interesting discoveries. Nested Reg-UNet architecture is found to outperform five other state-of-the-art architectures in both damage and component segmentation tasks. The architecture is built by combining a Nested UNet style dense configuration with a pretrained RegNet encoder. In terms of the mean Intersection over Union (mIoU) metric, the Nested Reg-UNet architecture provides an improvement of 2.86% on the damage segmentation task and 1.66% on the component segmentation task compared to the state-of-the-art UNet architecture. Furthermore, it is demonstrated that incorporating the Lovasz-Softmax loss function to counter class imbalance can boost performance by 3.44% in the component segmentation task over the most employed alternative, weighted Cross Entropy (wCE). Finally, weighted softmax ensembling is found to be quite effective when used synchronously with the Nested Reg-UNet architecture by providing mIoU improvement of 0.74% in the component segmentation task and 1.14% in the damage segmentation task over a single-architecture baseline. Overall, the best mIoU of 92.50% for the component segmentation task and 84.19% for the damage segmentation task validate the feasibility of these techniques for autonomous bridge component and damage segmentation using RGB images.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.365-381
• /
• 2023

The existing vision-based techniques for inspection and condition assessment of civil infrastructure are mostly manual and consequently time-consuming, expensive, subjective, and risky. As a viable alternative, researchers in the past resorted to deep learning-based autonomous damage detection algorithms for expedited post-disaster reconnaissance of structures. Although a number of automatic damage detection algorithms have been proposed, the scarcity of labeled training data remains a major concern. To address this issue, this study proposed a semi-supervised learning (SSL) framework based on consistency regularization and cross-supervision. Image data from post-earthquake reconnaissance, that contains cracks, spalling, and exposed rebars are used to evaluate the proposed solution. Experiments are carried out under different data partition protocols, and it is shown that the proposed SSL method can make use of unlabeled images to enhance the segmentation performance when limited amount of ground truth labels are provided. This study also proposes DeepLab-AASPP and modified versions of U-Net++ based on channel-wise attention mechanism to better segment the components and damage areas from images of reinforced concrete buildings. The channel-wise attention mechanism can effectively improve the performance of the network by dynamically scaling the feature maps so that the networks can focus on more informative feature maps in the concatenation layer. The proposed DeepLab-AASPP achieves the best performance on component segmentation and damage state segmentation tasks with mIoU scores of 0.9850 and 0.7032, respectively. For crack, spalling, and rebar segmentation tasks, modified U-Net++ obtains the best performance with Igou scores (excluding the background pixels) of 0.5449, 0.9375, and 0.5018, respectively. The proposed architectures win the second place in IC-SHM2021 competition in all five tasks of Project 2.

• Steel and Composite Structures
• /
• v.48 no.4
• /
• pp.367-383
• /
• 2023

A coupled wall consists of two or more reinforced concrete (RC) shear walls (SWs) connected by RC coupling beams (CBs) or steel CBs (hybrid-coupled walls). To fill the gap in the literature on the plastic hinge length of coupled walls, including coupled and hybrid-coupled shear walls, a parametric study using experimentally validated numerical models was conducted considering the axial stress ratio (ASR) and coupling ratio (CR) as the study variables. A total of sixty numerical models, including both coupled and hybrid-coupled SWs, have been developed by varying the ASR and CR within the ranges of 0.027-0.25 and 0.2-0.5, respectively. A detailed analysis was conducted in order to estimate the ultimate drift, ultimate capacity, curvature profile, yielding height, and plastic hinge length of the models. Compared to hybrid-coupled SWs, coupled SWs possess a relatively higher capacity and curvature. Moreover, increasing the ASR changes the walls' behavior to a column-like member which decreases the walls' ultimate drift, ductility, curvature, and plastic hinge length. Increasing the CR of the coupled SWs increases the walls' capacity and the risk of abrupt shear failure but decreases the walls' ductility, ultimate drift and plastic hinge length. However, CR has a negligible effect on hybrid-coupled walls' ultimate drift and moment, curvature profile, yielding height and plastic hinge length. Lastly, using the obtained results two equations were derived as a function of CR and ASR for calculating the plastic hinge length of coupled and hybrid-coupled SWs.

• Earthquakes and Structures
• /
• v.25 no.3
• /
• pp.199-208
• /
• 2023

Seismic performance analysis is one of the fundamental steps in the design of new or retrofitting buildings. In the seismic performance analysis, the adapted spectral acceleration curve for a given site mainly governs the seismic behavior of buildings. Since every soil site (class) has a different impact on the spectral accelerations of input motions, different spectral acceleration curves have to be involved for every soil class that the building is located on top of. Modern seismic design codes (e.g., Eurocode 8, EC8, or Turkish Building Earthquake Code, TBEC) provide design response spectra for all the soil classes to be used in the building design or retrofitting. This research aims to evaluate the EC8 and TBEC based design response spectra using the spectra of real earthquake input motions that occurred (and were recorded at only soil classes A, B and C, no recording is available at soil class D) in a specific area in Turkey. It also conducts response spectrum analyses of 5, 10 and 13 floor reinforced concrete building models under EC8, TBEC and actual spectral response curves. The results indicate that the EC8 and especially TBEC given design response spectra cannot be able to represent the mean actual spectral acceleration curves at soil classes A, B and C. This is particularly observed at periods higher than 0.3 s, 0.42 s and 0.55 s for the TBEC design response spectra, 0.54 s, 0.65 s and 0.84 s for the EC8 design response spectra at soil classes A, B and C, respectively. This is also reflected to the shear forces of three building models, as actual spectral acceleration curves lead to the highest shear forces, followed by the shear forces obtained from EC8 and, then, the TBEC design response spectra.

• Geomechanics and Engineering
• /
• v.30 no.6
• /
• pp.525-538
• /
• 2022

This paper presents a more general model for T-shaped combined footings that support two columns aligned on a longitudinal axis and each column provides an axial load and two orthogonal moments. This model can be applied to the following conditions: (1) without restrictions on its sides, (2) a restricted side and (3) two opposite sides restricted. This model considers the linear soil pressure. The recently published works have been developed for a restricted side and for two opposite sides restricted by Luévanos-Rojas et al. (2018a, b). The current model considers the uniform pressure distribution because the position of the resultant force coincides with the center of gravity of the surface of the footing in contact with the soil in direction of the longitudinal axis where the columns are located. This paper shows three numerical examples. Example 1 is for a T-shaped combined footing with a limited side (one column is located on the property boundary). Example 2 is for a T-shaped combined footing with two limited opposite sides (the two columns are located on the property boundary). Example 3 is for a T-shaped combined footing with two limited opposite sides, one column is located in the center of the width of the upper flange (b₁/2=L₁), and other column is located at a distance half the width of the strip from the free end of the footing (b₂/2=b-L₁-L). The main advantage of this work over other works is that this model can be applied to T-shaped combined footings without restrictions on its sides, a restricted side and two opposite sides restricted. It also shows the deficiencies of the current model over the new model.

• Advances in concrete construction
• /
• v.14 no.3
• /
• pp.169-183
• /
• 2022

In this article, vibrational behavior and wave propagation characteristics in (FG) functionally graded plates resting on Kerr foundation with three parameters is studied using a 2D dimensional (HSDT) higher shear deformation theory. The new 2D higher shear deformation theory has only four variables in field's displacement, which means has few numbers of unknowns compared with others theories. The shape function used in this theory satisfies the nullity conditions of the shear stresses on the two surfaces of the FG plate without using shear correction factors. The FG plates are considered to rest on the Kerr layer, which is interconnected with a Pasternak-Kerr shear layer. The FG plate is materially inhomogeneous. The material properties are supposed to vary smoothly according to the thickness of the plate by a Voigt's power mixing law of the volume fraction. The equations of motion due to the dynamics of the plate resting on a three-parameter foundation are derived using the principle of minimization of energies; which are then solved analytically by the Navier technique to find the vibratory characteristics of a simply supported plate, and the wave propagation results are derived by using the dispersion relations. Perceivable numerical results are fulfilled to evaluate the vibratory and the wave propagation characteristics in functionally graded plates and some parameters such wave number, thickness ratio, power index and foundation parameters are discussed in detail.

• Geomechanics and Engineering
• /
• v.32 no.1
• /
• pp.69-84
• /
• 2023

This paper proposes a novel frame element on Winkler-Pasternak foundation for analysis of a non-ductile reinforced concrete (RC) member resting on foundation. These structural members represent flexural-shear critical members, which are commonly found in existing buildings designed and constructed with the old seismic design standards (inadequately detailed transverse reinforcement). As a result, these structures always experience shear failure or flexure-shear failure under seismic loading. To predict the characteristics of these non-ductile structures, efficient numerical models are required. Therefore, the novel frame element on Winkler-Pasternak foundation with inclusion of the shear-flexure interaction effect is developed in this study. The proposed model is derived within the framework of a displacement-based formulation and fiber section model under Timoshenko beam theory. Uniaxial nonlinear material constitutive models are employed to represent the characteristics of non-ductile RC frame and the underlying foundation. The shear-flexure interaction effect is expressed within the shear constitutive model based on the UCSD shear-strength model as demonstrated in this paper. From several features of the presented model, the proposed model is simple but able to capture several salient characteristics of the non-ductile RC frame resting on foundation, such as failure behavior, soil-structure interaction, and shear-flexure interaction. This confirms through two numerical simulations.

• Earthquakes and Structures
• /
• v.22 no.5
• /
• pp.461-473
• /
• 2022

Energy-saving block and invisible multiribbed frame composite wall (EBIMFCW) is an important shear wall, which is composed of energy-saving blocks, steel bars and concrete. This paper conducted seismic performance tests on six 1/2-scale EBIMFCW specimens, analyzed their failure process under horizontal reciprocating load, and studied the effect of axial compression ratio on the wall's hysteresis curve and skeleton curve, ductility, energy dissipation capacity, stiffness degradation, bearing capacity degradation. A formula for calculating the peak bearing capacity of such walls was proposed. Results showed that the EBIMFCW had experienced a long time deformation from cracking to failure and exhibited signs of failure. The three seismic fortification lines of the energy-saving block, internal multiribbed frame, and outer multiribbed frame sequentially played important roles. With the increase in axial compression ratio, the peak bearing capacity and ductility of the wall increased, whereas the initial stiffness decreased. The change in axial compression ratio had a small effect on the energy dissipation capacity of the wall. In the early stage of loading, the influence of axial compression ratio on wall stiffness and strength degradation was unremarkable. In the later stage of loading, the stiffness and strength degradation of walls with high axial compression ratio were low. The displacement ductility coefficients of the wall under vertical pressure were more than 3.0 indicating that this wall type has good deformation ability. The limit values of elastic displacement angle under weak earthquake and elastic-plastic displacement angle under strong earthquake of the EBIMFCW were1/800 and 1/80, respectively.

• Structural Monitoring and Maintenance
• /
• v.10 no.1
• /
• pp.87-105
• /
• 2023

Seismic regulations have been updated from time to time to accommodate an increase in seismic hazards. Comparison of seismic fragility of the existing bridges in Indonesia from different historical periods since the era before 1990 will be the basis for seismic assessment of the bridge stock in Indonesia, most of which are located in earthquake-prone areas, especially those built many years ago with outdated regulations. In this study, seismic fragility curves were developed using incremental non-linear time history analysis and more holistically according to the actual strength of concrete and steel material in Indonesia to determine the uncertainty factor of structural capacity, βc. From the research that has been carried out, based on the current seismic load in SNI 2833:2016/Seismic Map 2017 (7% probability of exceedance in 75 years), the performance level of the bridge in the era before SNI 2833:2016 was Operational-Life Safety whereas the performance level of the bridge designed with SNI 2833:2016 was Elastic - Operational. The potential for more severe damage occurs in greater earthquake intensity. Collapse condition occurs at As = F_PGA x PGA value of bridge Era I = 0.93 g; Era II = 1.03 g; Era III = 1.22 g; Era IV = 1.54 g. Furthermore, the fragility analysis was also developed with geometric variations in the same bridge class to see the effect of these variations on the fragility, which is the basis for making bridge risk maps in Indonesia.