Corrosion poses significant challenges to steel structures, making early detection essential to avoid high maintenance costs and potential disasters. Automated detection and extent classification enable precise damage assessment, aiding targeted repair strategies and efficient resource allocation. Traditionally, human surveyors have conducted inspections, but these manual methods are labour-intensive and time-consuming and may yield inconsistent results. To overcome the conventional limitations, this study proposes applying image processing and a convolutional neural network (CNN) to corrosion detection, using a database of 9920 images captured with regular, portable cell phone cameras. The method uses 15 well-known pre-trained models (AlexNet, DenseNet-121, EfficientNet-B0, GoogleNet, Inception-v3, MobileNet-v1, NASNet, ResNet-18, ResNet-50, ResNet-101, ShuffleNet-v1, SqueezeNet, VGG16, VGG19 and Xception). CNN models are used to detect corrosion at ten levels, based on the percentage extent of corrosion, with levels ranging from 0% to 10% (level 1) and from 90% to 100% (level 10). After comparing 15 pre-trained models, Xception was chosen as the best, achieving a mean accuracy of 0.72, an accuracy of 0.93 and F1 and recall values of 0.91 and 0.94, respectively. This automated approach can aid early detection of corrosion, enhance maintenance prioritisation and reduce inspection costs for steel structures.
To address fatigue failure, weld cracking and resonance in steel-frame pedestals caused by mechanical vibration, this study proposes a composite steel frame fabricated by integrating concave hexagonal negative-Poisson’s-ratio honeycomb structures with I-shaped steel, aiming to achieve the integrated structural function of load bearing and vibration damping. Bending and vibration tests are carried out on both conventional and composite steel frames to compare their static load-bearing characteristics and dynamic vibration-damping effects. Meanwhile, a finite-element model is established based on the Abaqus software platform to explore the regulation mechanism of honeycomb geometric parameters on the composite frame’s performance. Results show the composite frame realises ‘load-bearing–vibration-damping’ synergy, with yield strength basically consistent with conventional frames and excellent low-frequency vibration energy dissipation capacity. Honeycomb arrangement position and cell thickness mainly regulate load-bearing performance, while layer number and cell thickness have a significant impact on damping. After parameter optimisation, the maximum vibration level difference of the composite frame peaks at 62.81 dB, providing new design ideas and technical support for steel frame performance optimisation under low-frequency vibration environments.
This paper presents a comprehensive architectural and structural assessment of the Tekirdağ Archaeology and Ethnography Museum, a heritage building characterised by a mixed structural system consisting of rubble masonry and reinforced concrete additions. Unlike many previous studies focusing primarily on religious or purely masonry structures, this research investigated a museum building where both structural safety and the protection of cultural artefacts are critically important. The methodology integrated architectural surveys, in situ damage assessment, soil characterisation based on multi-channel analysis of surface waves, material testing and three-dimensional finite-element modelling using Lusas. The results showed that the structure is stable under gravity loads, but seismic analysis revealed tensile stresses exceeding material capacity, indicating a high risk of damage during potential earthquakes. The originality of this study lies in combining field data, geotechnical analysis and numerical modelling within a single framework applied to a museum structure. The findings highlight the necessity of interdisciplinary evaluation approaches and provide practical recommendations for improving seismic performance while preserving heritage value.
This study develops refined closed-form analytical solutions for accurately determining local buckling coefficients and corresponding critical local buckling stresses of rectangular hollow sections (RHSs) with web-dominant thickness under axial compression. A novel local buckling model is proposed using a reference-section analytical framework that establishes geometric and mechanical equivalence between an original unequal-thickness RHS and two reference RHSs with equal wall thickness. The first reference section adopts the flange thickness, while the second represents the web thickness, enabling systematic characterisation of local buckling behaviour. Explicit analytical expressions are derived by introducing corner-size factors to capture coupled plate interaction effects and by enforcing compatibility between the buckling responses of the reference and original sections. The proposed solutions are validated against linear elastic local buckling simulations conducted in the Abaqus software program, showing minimal deviations. Results demonstrate that the model accurately predicts local buckling coefficients and stresses, providing slightly conservative yet reliable estimates. Comparisons with previous energy-based studies show close agreement, whereas methods relying on indirectly derived coefficients tend to overestimate stresses. The originality of this work lies in presenting explicit closed-form solutions for unequal-thickness RHSs with web-dominant thickness, addressing a significant gap in the existing literature.
In this paper, the advantages and limitations of certain methods are evaluated in terms of theory and their practical implementation, when considering the analysis of damage diagnosis methodologies in civil engineering structures in relation to new structural dynamic properties. Methods of damage diagnosis such as reliance on natural frequency, vibration mode, a change of the finite-element model and the flexibility matrix are discussed in detail in this paper. The results indicate that the reconstruction error decreases by a maximum of 15% with increase in damage severity, and the maximum damage index has a growing trend. The accuracy of the proposed method in detecting the location of multiple instances of damage reaches more than 90%, indicating a high level of consistency in different damage cases. In the experiment conducted, the peak value of the single-damage index curve is clearly visible at the point of damage, and the peak value grows depending on the severity of the damage. Specifically, the variation of the reconstruction error changes by an amount of −15%, whereas the maximum value of the damage index increases in the presence of more severe damage. It can be stated that the technique developed exhibits great reliability in detecting damage under varying conditions. In particular, over 90% accuracy was achieved when determining multiple damage points. Moreover, although different severities of damage were present at some points, all damaged positions were detected successfully.
This research examined the hypothesis of the existence of a force due to acceleration changes in structures. The existence of a jerk force was first mathematically proven in the presence of initial acceleration conditions. By defining a third-order equation that has the capability to consider the jerk force, its solution was derived. The solution of this differential equation was performed both directly with initial conditions and through generalisation of Duhamel's integral. Comparison of the solutions of second- and third-order equations with experimental results using the proportional jerk coefficient showed good agreement. The greatest differences were when the ratio of loading frequency to the natural frequency of the structure, considering the damping ratio, was greater than 6.75. The effect of earthquakes as loads on structures, considering the ground jerk effect, showed that, in the third-order equation, the acceleration of tall or damaged structures with a small natural frequency and high damping was greater than in the second-order equation. Due to the out-of-phase nature of acceleration and jerk, the maximum values of these two do not occur together. The ratio of maximum jerk to maximum acceleration is an important variable in calculating the jerk effect.
Damage detection of civil engineering structures using the vibration-based approach is affected by the environmental and operational conditions the structures face. This is because these conditions also affect the vibration properties of the structures, which are commonly analysed to detect damage. Moreover, the presence of outlier measurements also affects the performance of the damage detection methods. Outliers are measurements with abnormal values, and they can create masking effects where small levels of damage are not identified. Therefore, a method is proposed in this paper to detect damage under the effects of environmental and operational conditions, and outlier effects. It is proposed to analyse the regression coefficients of the regression model of the database of vibration properties obtained from a structure for damage detection. The regression coefficients are sensitive to the presence of abnormalities in the database, which can be due to damage and outliers. To test the proposed method, an experimental wooden bridge is analysed, and the results obtained demonstrate that damage can be detected under the influence of environmental and operational conditions, and outlier effects.
To improve construction efficiency and maintenance access in steel-concrete composite bridges, in this study, a prefabricated system is proposed using cold-formed boolean AND-shaped steel girders and lightweight aggregate concrete (LWCC) slabs. Four-point bending tests were conducted to compare the flexural performance of the proposed beam against a conventional composite beam. Key parameters, including failure modes, load capacity, stiffness and the performance of ultra-high-performance concrete (UHPC) wet joints with an anchorage length of 10 (where is the bar diameter) were investigated. Experimental results show that the boolean AND-shaped steel-LWCC system achieves competitive structural performance. Specifically, the cracking load increased significantly by 31.7%, while ultimate load and flexural stiffness showed modest improvements (approximately 4%) compared to the conventional beam. Minimal interfacial slip confirmed reliable composite action, and the boolean AND-shaped configuration effectively reduced shear lag. The joints remained intact, ensuring monolithic behaviour. The system's practical feasibility was validated through successful implementation in a bridge project. This study demonstrates that the novel system offers enhanced constructability, reduced self-weight and improved inspectability for prefabricated bridge applications.
This discussion identifies possible inconsistencies in the fundamental modelling parameters and input data used in the study by Dolati et al. (2023). It reveals that the finite element calibration relied on simplified assumptions, specifically the use of a constant volumetric ratio (𝛽 = 0.2) across varying confinement levels and a generous softening parameter (𝑤 𝑑 = 50 𝑚𝑚). When combined with variations in the unloading factor (𝑓 𝑈 = 0.3 𝑎𝑛𝑑 0.6), these assumptions may lead to simulated responses that differ from the actual behavior observed in real-life columns.
Research on recycled aggregate concrete () has grown in response to the need for sustainable management of construction and demolition waste. However, applications in partially encased composite beams (PECBs) remain limited. This study investigates the flexural behaviour of PECBs with through material characterisation, a four-point bending test and numerical modelling using the Abaqus v6.16. A parametric analysis assessed the effects of steel profile geometry, concrete type and steel yield strength on the load capacity of the beam. Concrete mixtures with 0%, 30%, 50% and 100% natural coarse aggregate replacement were characterised, and one with 50% replacement (RAC-50) was tested in bending. The RAC-50 beam reached a peak load about 5% higher than the natural aggregate concrete specimen, showing greater ductility, slightly larger deflections and similar failure modes. Numerical results reproduced peak loads within 5% error and indicated that replacement up to 50% has minimal influence on flexural behaviour. Parametric studies showed that steel profile height was the dominant geometric factor, affecting capacity by up to 74%, while yield strength influenced it by up to 18%. Overall, the results highlight the structural viability of using in PECBs and support its adoption in sustainable composite construction.
Corrosion is a major cause of deterioration in steel-reinforced concrete structures. Existing models present two major challenges: they require detailed corrosion data that are not readily accessible during non-destructive inspections, and many are derived from limited data sets, which compromises their predictive accuracy when applied at a larger scale. This study proposes a simple regression-based model that utilises easily measurable simple parameters (specifically crack width and cover-to-diameter ratio) as primary inputs to estimate the residual reinforcement area, supporting preliminary assessment of reinforced concrete (RC) structures. A database of 465 experimental data points was compiled and reduced from 17 to six parameters. Sensitivity analysis identified the crack width and cover-to-diameter ratio as the primary input parameters of the developed model. This model predicts the penetration depth and the corresponding residual bar diameter. Comparative evaluation against existing models demonstrated better predictive accuracy. The proposed model provides a practical tool for structural engineers to estimate the remaining uncorroded bar area and obtain a preliminary upper-bound estimate of the residual load-carrying capacity of RC members for preliminary assessment. The model assumes uniform corrosion and does not account for localised pitting, bond degradation or ductility reduction, and therefore is not intended to replace a detailed structural assessment.
This study aims to elucidate the synergistic action of steel and polyvinyl alcohol (PVA) hybrid fibres in glass fibre-reinforced polymer (GFRP)-reinforced concrete columns and to develop a reliable prediction model for axial compressive capacity. Axial compression tests were performed on nine specimens: eight hybrid fibre-reinforced GFRP concrete columns with varied fibre dosages and one conventional steel-reinforced control column. Failure modes, load–displacement behaviour and strain responses of longitudinal reinforcement, transverse reinforcement and concrete were compared across mixtures. Fibre-free GFRP columns exhibited sudden, brittle failure. Steel fibres primarily enhanced ultimate capacity; at 1.4% and 0.8% volume fractions, ultimate loads increased by 17.1% and 13.7%, respectively, relative to the steel control and by 24.7% and 21.1%, respectively, relative to the fibre-free column. However, excessive steel fibre content (>1.4%) caused poor dispersion and weaker fibre–matrix bonding, reducing reinforcement efficiency, axial displacement at peak load and overall ductility. In contrast, PVA fibres improved post-peak deformation and energy absorption, and hybridisation provided stable, balanced performance gains. Based on the experimental data set, relevant design codes and micromechanical interpretation, an optimised ultimate-capacity formula is proposed, showing good agreement with the measured results. The model offers a practical basis for design of hybrid-fibre GFRP columns.
Ultra fibre-reinforced concrete (UFRC) layers were used to strengthen two-way concrete slabs and their stiffness under monotonic four-point bending was evaluated. Separated into three groups with low, medium and high reinforcement ratios (groups W, M and S, respectively), a total of 15 slabs were cast and tested. N each group, one slab was not strengthened, while the rest were reinforced using UFRC layers containing 1% or 2% polypropylene fibres. These UFRC layers were applied either on the tensile side of the slab or on both tensile and compressive sides. The flexural behaviour, crack development and failure modes were examined. The results showed that the cast-in-place UFRC layers significantly improved the yield load, ultimate load and stiffness, while also reducing the final deflection. They also enhanced the ductility, energy absorption capacity and cracking strength of the slabs. Notably, displacement ductility increased in all slabs strengthened with 2% fibres and UFRC sheets on two sides. Compared with the respective control specimens, the ductility increased by factors of 2.75, 1.50 and 1.71 in groups W, M and S, respectively. Overall, the findings demonstrate that UFRC layers provide an effective strengthening technique for deficient slabs, offering improvements in load capacity, deformation resistance and overall structural performance.
This study examines using heavy-duty metal straps bolted onto the face of concrete beams as an approach of reusing discarded heavy-duty metals to strengthen concrete beams. The method is quick, cost-effective and time-saving, and provides a practical solution for strengthening beams regardless of their casting methods or locations within a building. A set of 12 reinforced concrete beams was constructed to fail under bending forces. The beams were designed with a width of 160 mm, height of 240 mm and a total length of 2100 mm. To attach the metal straps, 14 mm holes were drilled at different locations of the concrete beams and M12 thru bolts were used to secure the metal straps. The testing parameters were the number of the straps and their positions on the face of the beam. It is found that bolting metal straps on the beams enhances the load-carrying capacity of the beams by up to 70%. To achieve the optimal strength, it is essential to ensure that the holes are precisely at the centre of the strap and the straps are aligned with the provided steel reinforcement of the beam. Finally, the behaviour of the beams is validated by VecTor2 software.
In this study, the seismic performance of buildings with coupled irregularities is investigated – both in plan and elevation; such buildings are common in hilly terrains. A field survey is conducted using rapid visual screening to investigate prevalent building configurations in hilly regions, with an emphasis on Srinagar Garhwal, Pauri town and Rudraprayag city located in the state of Uttarakhand, India. In total, 1558 structures exhibiting plan and/or vertical irregularities are documented, influenced by the underlying slope gradient. Based on the identified configurations, 12 building models incorporating re-entrant corners, non-parallel lateral force systems, step-back and split foundations are developed in finite-element-based software. Non-linear time history analysis is performed using a suite of 11 ground motions. Key structural parameters, including the fundamental period, torsional irregularity, interstorey drift, peak roof displacement and peak floor acceleration, are evaluated. The results highlight that buildings with coupled irregularities experience significantly amplified seismic demands in contrast to the regular models. The study also investigates the influence of seismic incidence angle, identifying 135° as the most critical. Unlike prior studies, this work addresses a critical gap in code-based design practices and provides valuable recommendations for seismic design guidelines, emphasising the need for irregularity-sensitive detailing in hillside construction.
In this study, the mechanical behaviour of bolted timber joints with slotted-in steel plates was investigated experimentally, comparing flat plates with innovative corrugated plates, both with and without carbon fibre-reinforced polymer () reinforcement. Thirty-two specimen sets (16 corrugated, 16 flat) were tested under monotonic loading parallel to the grain using an orthogonal design. Results showed that corrugated-plate joints exhibited distinct failure modes including shear, compression and transverse-longitudinal () cleavage, with combined bolt yield deformations. Their load-displacement curves lacked a clear yield plateau, progressing through elastic, fracture-transition and failure stages. Corrugated configurations demonstrated superior stiffness and bearing capacity compared to flat-plate joints. U-shaped reinforcement significantly enhanced corrugated-plate joint performance, an effect absent in flat-plate joints. Theoretical analysis based on the Eurocode revealed that current formulas are applicable to flat-plate joints but overlook end-region mechanics in corrugated-plate joints. A simplified empirical method incorporating timber end characteristics and error analysis was proposed. However, mechanical aspects such as fracture effects require further investigation.
Smart coatings are advanced functional materials designed to respond adaptively to environmental stimuli such as temperature, pH, light and mechanical stress. Unlike conventional coatings, they provide dynamic functionalities including self-healing, anti-microbial activity, self-cleaning and environmental responsiveness, making them essential for applications in aerospace, biomedicine, energy and electronics. Recent progress in multifunctional smart coatings is summarised in this review, with a focus on classification based on functional behaviour such as self-healing, stimuli-responsive, anti-microbial and conductive systems. The underlying mechanisms are discussed, including microencapsulation, dynamic covalent bonding, supramolecular interactions and responsive polymer networks. Fabrication techniques such as sol–gel processing, layer-by-layer assembly, electrochemical deposition and vapour deposition are evaluated in terms of structural control and performance optimisation. Industrial applications are highlighted and current challenges related to durability, scalability and environmental safety are identified. Future perspectives emphasise sustainable material design, multifunctional integration and digital connectivity for intelligent surface systems.
In the swivel construction of large-tonnage cable-stayed bridges, the swivel speed often plays an important role in stability and safety. Based on the background of the largest swivel bridge across Xiangyang North Marshalling Station in Asia, this paper explores the influence of different rotational speeds on the response of swivel construction through model testing and finite-element analysis. The results of the model test and the finite-element analysis are compared and analysed. The results show that with the increase of the rotational speed, the fluctuation range of the vertical displacement and stress of the main beam of the bridge increases, indicating that the higher rotational speed will lead to a stronger dynamic response of the bridge structure.
Quasi-static investigations have demonstrated that continuous reinforcement traversing the columns enhances the load-bearing capacity of reinforced concrete (RC) flat slab systems subjected to progressive collapse initiated by column removal. Nevertheless, the dynamic behaviour of older structures lacking reinforcement remains insufficiently understood. To address this gap, high-fidelity numerical models were developed in LS-Dyna and validated against experimental data. In addition, the load redistribution and internal force transfer mechanisms were examined in this study. The results revealed that, following the removal of a central column, inertial effects amplified the applied load by a factor of 1.12, with more than 60% of the load ultimately redistributed to the edge columns. The influence of varying column removal scenarios and structural scales on the dynamic response and collapse resistance of flat slab systems was also explored. Findings indicate that the simultaneous loss of opposite edge columns imposes a greater collapse risk than the concurrent loss of adjacent edge columns. Moreover, the evaluation results of scaled RC flat slab structures tend to overestimate both their dynamic load-bearing capacity and punching shear resistance.
