This study systematically investigates the flexural buckling performance of welded I-section beam-columns fabricated from Q1100 ultra-high strength steel. Beam-column tests were conducted on six pin-ended specimens susceptible to minor-axis buckling, providing critical mechanical response data. The experimental results were used to validate a finite element model, which was subsequently employed for extensive parametric analysis. Integrating experimental and numerical findings, the applicability of current design codes—including those from Europe, North America, Australia, and China—was evaluated. Finally, an improved design approach is proposed within the Eurocode framework, using the Modified Direct Strength Method to calculate the bending bearing capacity of the section, and combining the interaction coefficients kz that have been calibrated for non-long and long sections respectively. Through experiments and finite element data verification, the applicability of this method is excellent.

Wire-and-arc additive manufacturing (WAAM) is a metal 3D printing technique that enables the production of complex large-scale components efficiently and cost-effectively, attracting attention from the construction industry. However, the existing knowledge on the low-cycle fatigue performance of WAAM stainless steel is scarce, hindering its engineering applications under seismic or wind loading. To this end, strain-controlled low-cycle fatigue tests were conducted on WAAM 316 L stainless steel coupons extracted from three directions. Machined coupons were tested as the primary basis for characterizing the intrinsic cyclic response and strain-life relationship of the material, while as-built coupons were examined to evaluate the influence of the tested surface morphology and geometric variability. A total of sixty fatigue specimens, covering two surface conditions, three extraction directions and five strain amplitudes, were tested. The machined coupons exhibited cyclic hardening and non-Masing behaviour, and their fatigue lives showed limited dependence on the extraction direction after surface irregularities were removed. Fractographic analysis revealed that fatigue cracks in machined coupons typically initiated from internal porosity or edge areas, whereas the rough surface was the primary factor for asbuilt coupons. The Basquin-Coffin-Manson and Kuroda models were employed to predict fatigue life with good accuracy. Compared with machined coupons, the tested as-built coupons exhibited shorter fatigue lives due to surface-induced stress concentration and geometric variability.

This study presents a systematic experimental and numerical investigation on uncoated Q235 steel plates. Steel plates were first subjected to neutral salt-spray exposure with four different durations to induce controlled corrosion. Subsequently, 3D scanning was performed to extract pit geometry descriptors such as depth and aspect ratio. Tensile specimens prepared from corroded plates were tested to evaluate mechanical degradation. The results show that pit depth and aspect ratio follow lognormal distributions, and that yield strength, ultimate strength, strain at ultimate tensile strength, and elastic modulus degrade approximately linearly with corrosion mass loss in nominal terms. Based on the statistical morphology parameters, finite element models were developed and validated against tensile test data. Parametric analyses covering mass loss ratios from 0.04 to 0.32 were conducted to enable full-range calibration of the degradation relationships, avoiding reliance on experiments alone. These analyses enabled the calibration of empirical degradation equations and supported the formulation of a three-stage constitutive representation that captures mass-loss-induced deterioration. Its performance was further evaluated through reliability analysis, conducted using the first-order reliability method in accordance with EN 1990 and based on the statistical variability of experimental data and FE predictions. The analysis yielded partial safety factors for key mechanical parameters, thereby facilitating the direct application of the proposed model in structural design.

In response to the growing demand for sustainable and low-carbon building systems, this study presents a comprehensive experimental investigation on the axial compression behavior and composite interaction characteristics of stainless steel-timber composite (SSTC) columns. A total of 14 specimens were fabricated, covering four cross-sectional configurations and two interfacial connection methods (self-tapping screws and epoxy adhesive), with pure stainless steel and pure timber columns of identical dimensions serving as control specimens. The load-axial displacement, lateral displacement, rotation, strain responses and failure modes were analyzed to quantify the initial stiffness, ultimate load-bearing capacity, and ductility. The composite efficiencies of different section types and connection methods were also compared. Based on the superposition method, a calculation formula for the axial load-bearing capacity of SSTC columns was established, and the stability factors specified in different design standards were adopted for capacity prediction.

Accurately identifying parameters for phase-field fracture models is crucial for predicting material failure but remains challenging due to the ill-posed nature of the inverse problem, where different parameter combinations can yield nearly identical mechanical responses. To address this non-uniqueness issue, this study proposes a novel cycle-consistent Physics-Regularized Neural Network for the robust calibration of constitutive and phase-field fracture model parameters. The framework establishes bidirectional mappings between the parameter domain and the load-displacement response domain using two interconnected neural networks. A forward cycle-consistency loss is introduced to enforce a closed-loop constraint, effectively regularizing the solution space and mitigating non-uniqueness. Furthermore, physical constraints derived from governing equations and a specialized penalty on the critical fracture energy parameter are embedded to enhance prediction accuracy. The model is validated against baseline and ablated variants, demonstrating that the integration of cycle consistency and physical knowledge significantly improves parameter estimation and response curve reconstruction. The model's generalizability is further confirmed by accurately predicting fracture behavior across various specimens with complex stress states not included in the training dataset, showcasing its robustness for inverse parameter identification in complex material mechanics.

Selective laser melting (SLM) of 316L stainless steel produces hierarchical microstructures that lead to complex cyclic deformation behavior. However, the macroscopic transition between cyclic softening and cyclic hardening, particularly under the influence of build orientation, remains insufficiently understood. In this study, the orientation-dependent low-cycle fatigue (LCF) behavior of stress-relieved SLM 316L was systematically investigated for three build orientations (0 degrees, 45 degrees, and 90 degrees) over a strain-amplitude range of 0.5 %-2.0 %. By analyzing the evolution of hysteresis energy dissipation (Wp), a strain-amplitude-dependent transition in cyclic behavior was identified: continuous softening dominates at low strain amplitudes, whereas pronounced secondary hardening appears at 2.0 %. In addition, an anomalous inversion of Wp was observed, in which the energy dissipation at 2.0 % became lower than that at 1.8 % for the 45 degrees and 90 degrees orientations. Post-fatigue XRD results obtained from gauge-section cross-sections indicate that the terminal BCC martensite fraction increases at high strain amplitudes, which is consistent with the observed secondary hardening tendency. These results suggest that build orientation affects the balance between recovery-related softening and transformation-assisted hardening, thereby changing hysteresis energy evolution and fatigue resistance. The present work provides a macroscopic basis for interpreting the orientation-dependent cyclic response of stress-relieved SLM 316L.

Orthotropic steel decks (OSDs) have been widely adopted in bridge structures due to their lightweight construction, high load-carrying capacity, and ease of fabrication. However, fatigue cracking poses a significant threat to their long-term durability. The fatigue vehicle models specified in current design codes are based on outdated traffic survey data and therefore do not accurately reflect current regional traffic characteristics. In this study, a measured fatigue vehicle model reflecting the traffic characteristics of the region was developed based on long-term weigh-in-motion (WIM) data collected from some bridge in Beijing. Full-scale fatigue tests were conducted using both the measured and code-specified vehicle models. Numerical simulations and fracture mechanics approach were further performed to systematically evaluate the fatigue behavior of OSDs. The results show that the code-specified fatigue vehicle model differs markedly from actual traffic conditions, whereas the measured model generates substantially higher fatigue stresses at critical OSD details. Fatigue life evaluation indicates that the fatigue life under measured vehicle loads is approximately 18 % shorter than that predicted by the code-specified model, demonstrating a potential safety risk in fatigue design when relying on outdated code vehicle models. Therefore, incorporating measured traffic load characteristics into bridge fatigue design and assessment is recommended to improve the accuracy of fatigue life predictions and ensure structural safety.

Wire arc additive manufacturing (WAAM) has significant potential in the construction industry owing to its high productivity, cost-effectiveness and high degree of automation. However, the understanding of active interlayer cooling (AIC) and its effects remains limited. To address this gap, three carbon steel plates were fabricated via WAAM using ER50-6 welding wire under natural cooling, active compressed air cooling and active water spray cooling, respectively. A total of 45 as-built and machined tensile coupons were tested to systematically investigate the effects of AIC strategies on geometric and stress–strain characteristics. The results show that the examined WAAM carbon steel exhibited a stress–strain response comparable to that of hot-rolled normal-strength carbon steels. The effective mechanical properties were influenced by the geometric topography; specifically, surface undulations introduced anisotropy and negatively affected mechanical properties, reducing the fracture strain by up to 18%. The AIC strategies effectively mitigated heat accumulation and significantly reduced interlayer dwell time, thereby increasing the productivity to up to 2.35 times that for natural cooling. In terms of mechanical properties, the average yield strengths under natural cooling, active compressed air cooling and active water spray cooling were 354.8, 354.7 and 357.3MPa, respectively; the tensile strengths under the two active cooling strategies were also comparable to those associated with natural cooling, whereas the fracture strain generally decreased, with a maximum reduction of 12%. These results demonstrate the effectiveness of the employed AIC strategies and the minimum impact on the resulting mechanical properties. Overall, the examined WAAM carbon steel with AIC exhibited consistent mechanical properties and enhanced production efficiency, satisfying the requirements for structural engineering applications.

This study investigates the residual patch loading resistance of corroded steel H-section beams through nonlinear finite element analysis. A morphology-informed corrosion modelling framework was developed to explicitly represent the realistic geometry and spatial distribution of corrosion pits, together with the associated nonuniform web thinning, thereby enabling a more realistic simulation of corrosion-induced local weakening than simplified uniform-thinning approaches. The numerical models were validated against available test results reported in the literature, and a parametric study comprising 432 simulations was then conducted to quantify the effects of corrosion severity, web slenderness, and bearing width ratio on patch loading resistance. The parametric analysis results show that corrosion morphology and web slenderness are the dominant factors governing resistance degradation. The equivalent uniform-thinning method was found to overestimate the patch loading resistance by approximately 8% when the mass-loss ratio reached 0.40. Increasing the web slenderness ratio from 21 to 75 led to an average reduction of approximately 78.6% in patch loading resistance. Comparisons with the design provisions in EN 1993-1-5 (2006), ANSI/AISC 360-22 (2022) and GB 50017 (2017) demonstrated that the current standards generally provide unconservative predictions for corroded H-section beams under patch loading. To address this limitation, a corrosion-induced reduction factor was therefore proposed to improve the prediction accuracy of the residual patch loading resistance.

High-strength steel bars in harsh environments suffer uneven corrosion that impairs seismic performance. To investigate the corrosion characteristics, degradation mechanism of post-corrosion tensile properties, and prediction method of high-strength steel bars, this study conducted accelerated corrosion tests under dry-wet cycles, three-dimensional laser scanning tests, three-dimensional digital image correlation (3D-DIC) measurements, and monotonic tensile tests on 33 steel bar specimens with varying degrees of corrosion, material strengths, and diameters. A two-component Gaussian mixture model fitted to scanning data effectively predicted residual cross-sectional area distribution (with a mean R2 value exceeding 0.85), enabling pit severity assessment via component means and weights. Fracture patterns at ultimate load were analyzed using 3D-DIC and residual area. Test-derived stress-strain curves show that characteristic strength decreased linearly with corrosion rate, with smaller diameters and higher strengths suffering greater attenuation. The prediction model for the tensile properties of corroded steel bars was established using multiple methods, including refined finite element simulation, explicit constitutive equations, and a fully connected neural network (FCN). This paper enriches the corroded rebar property database and provides a versatile prediction tool applicable to various scenarios.

SLM 316L stainless steel contains a hierarchical microstructure, including columnar grains, EBSD-resolved low-angle boundary networks, and crystallographic texture. How these features jointly control early plastic anisotropy remains unclear. This study investigates the orientation-dependent deformation of stress-relieved SLM 316L using tensile testing and semi-in-situ EBSD. Miniature dog-bone specimens with tensile axes of 0°, 45° and 90° were interrupted at 0%, 2% and 10% engineering strain. Kernel average misorientation, geometrically necessary dislocation density, grain orientation spread, and Schmid-factor-based slip-tendency descriptors were analyzed to separate crystallographic slip tendency from morphological constraint. The 45° specimen showed the highest stress level during early yielding, followed by the 0° specimen, whereas the 90° specimen exhibited the lowest resistance to plastic flow. At 2% strain, the 45° specimen developed the largest increase in median GND density and high-GND area fraction, although the 90° specimen showed the highest maximum Schmid factor, mmax, and the strongest multi-slip tendency. This mismatch shows that crystallographic slip tendency alone cannot explain early strain partitioning. From 2% to 10% strain, the 45° specimen also showed the largest increases in GND density and GOS, indicating sustained lattice-curvature accumulation and grain-scale orientation dispersion. Regional analysis further showed that LAGB-associated regions preferentially accumulated GND and KAM at 2%, while this contrast weakened at 10%. These results reveal that early plastic anisotropy in SLM 316L is governed by the coupling between crystallographic slip tendency, columnar morphological texture and retained low-angle boundary network.

Damage in stainless steel frame connections under seismic cyclic loading, especially weld cracking and bolt slippage, may occur before obvious global stiffness degradation, making early connection-level structural health monitoring (SHM) challenging. This study applies acoustic emission (AE) technology to characterize damage modes in stainless steel frames with bolted-welded hybrid and extended end-plate connections. AE multi-parameter features, time-frequency characteristics, spectral energy ratios, load and calm ratios, and RA-AF clustering were jointly analyzed to compare the AE responses associated with weld cracking and bolt slippage. A representative weld cracking signal was dominated by low-frequency components within the 0 to 100 kHz range, with an energy proportion of 95.3%. A representative bolt slippage signal exhibited a distinct spectral peak at 173.8 kHz and a characteristic energy component in the 150 to 200 kHz band, with an energy proportion of 30.8%. AE activity associated with bolt slippage was detected at an interstory drift ratio of 0.75%, earlier than the approximately 2.0% drift level at which stiffness degradation became observable in the hysteresis curves. Furthermore, the load and calm ratios, together with RA-AF clustering, characterized the transition from relatively stable AE behavior to progressive plastic deformation, end-plate opening, and slippage accumulation. These results provide experimental evidence for AE-based identification and early warning of weld cracking and bolt slippage in stainless steel frame connections, and offer useful guidance for connection-level SHM.

This paper investigates the structural behaviour and design of stainless-steel (SS) T-stubs connected by swagelocking pins. Sixteen swage-locking pinned SS T-stubs were tested under monotonic tension. The experimental results, including failure mode, ultimate resistance, deformation capacity and load-displacement responses, were reported. Preload measurement and tensile tests on individual swage-locking pins were additionally conducted to evaluate their preload stability and tensile resistance. Finite-element (FE) models for both T-stubs and swagelocking pins were developed and validated against the experimental data. An extended parametric study was performed to investigate the effect of key parameters-pin pitch, preload, equivalent segment length, and pin diameter-on the structural behaviour of T-stubs. The existing design methods for predicting the resistance of SS T-stubs, including the design provisions in EN 1993-1-8 (EC3) and the Continuous Strength Method (CSM), were evaluated. The results show that EC3 significantly underestimates the ultimate resistance of this connection. The CSM is a deformation-based framework that accounts for stainless-steel strain hardening and provides a higher flange plastic resistance Mf,Rd. The results indicate that the CSM improves the prediction accuracy of ultimate resistance while remaining conservative. Therefore, a new CSM-based design method is proposed for stainlesssteel T-stubs connected by swage-locking pins. It combines the CSM-based flange resistance with an explicit consideration of the pin contribution. The method provides improved accuracy and consistency compared with EC3 and the CSM.

To address the conflict between high load-bearing capacity and smooth energy absorption in protective structures, this study proposes a novel bi-layer cascaded metamaterial consisting of a rigid Polylactic Acid (PLA) substrate and hard Thermoplastic Polyurethane (TPU) truncated conical shells. A full-scale three-dimensional numerical model was established to investigate its nonlinear response mechanism under axial compression. The simulation results reveal a distinct sequential buckling mechanism, which effectively mitigates the theoretical elastic peak load while maintaining a high plateau-to-peak force ratio. To accurately predict the mechanical response, a physically-based modified analytical model based on curved beam theory is developed. Considering the nonideal constraints in engineering applications, a boundary rotational stiffness correction factor [Formula: see text] is derived based on structural stability theory to account for the semi-rigid boundary effect caused by the synergistic deformation of the substrate. Furthermore, a delayed linear support term, activated at the onset of structural densification (approximately half of the shell height), is introduced to characterize the post-collapse stiffness dominated by substrate bending. The modified analytical predictions show good agreement with numerical results, validating the proposed mechanical framework. The structure exhibits a Specific Energy Absorption (SEA) of approximately 25[Formula: see text]J/g, demonstrating significant potential for impact protection applications requiring both force-limiting capabilities and high energy dissipation.

Wire arc additive manufacturing (WAAM) is a mature metal 3D printing technique with highly-automation, greater design freedom and large-scale printing capabilities. Applications of WAAM in construction have only emerged in recent years and it has been shown that the mechanical properties are significantly affected by printing process parameterization. In this study, an experimental study into the mechanical and high cycle fatigue (HCF) properties of WAAM ER70S-6 and ER110S-G steels was conducted. A total of 10 tensile coupons under static tensile loading and 31 HCF coupons under constant HCF amplitude loading were tested. Mechanical properties of WAAM steels at room temperature were determined and the axial strain distribution on the as-built surface was observed by digital image correlation (DIC). S-N curves of as-built normal-strength and high-strength steels were derived from test data and compared to the fatigue design curves in IIW Recommendations. Based on 3D laser scanning, geometric measurements of WAAM coupons were performed and the effect of local stress concentration on fatigue life was considered. Additionally, fractography of the tested coupons was performed to assess their failure mechanisms. Comparing two types of as-built coupons, the normal-strength WAAM steel provided less geometric variability and exhibited much better ductility compared to high-strength steel. Normal- strength WAAM steel exhibited better high cycle fatigue behavior compared to both former test results and high- strength steel. Severe stress concentration and multiple fracture mechanisms led to inferior HCF properties of ER110S-G steel. This study extends the WAAM fatigue experimental data pool for carbon steel materials and evaluates the effect of deposition strategies on the material properties, demonstrating promising application prospects in structural engineering.

This study employs acoustic emission (AE) technology to monitor the tensile tests of stainless steel weld specimens, analyzing and comparing the mechanical properties of the welds and base materials, as well as the AE characteristic differences during crack propagation. The results indicate that rise time (RT) is introduced for the first time as a quantitative indicator for identifying the initiation and propagation of early microcracks in welds. By integrating RT with the RA and AF parameters, the tensile process of welds can be divided into three distinct stages: elastic, strengthening, and fracture. Furthermore, a novel hypothesis for weld damage stages is proposed based on AE waveform characteristics. Additionally, the maximum load ratio (L/Lmax) is introduced as a quantitative indicator for the first time, revealing that the load-bearing capacity of welds during early microcrack formation is significantly lower than that of the base material. To further enhance damage characterization, wavelet transform is employed to extract the time-frequency and amplitude-frequency features of AE signals. An image dataset is constructed and utilized in the GoogLeNet model for intelligent identification of weld damage states. The model exhibits excellent convergence and high accuracy, thereby optimizing the completeness of feature extraction and overcoming the limitations of traditional methods in analyzing the damage evolution process.

The structural health monitoring (SHM) of large-scale concrete infrastructure is often impeded by the prohibitive cost of deploying dense sensor arrays for direct damage detection. This study's primary objective is to develop a cost-effective, intelligent framework to accurately infer local structural damage from sparse, global measurements. A novel "sparse-to-dense" methodology is proposed, which utilizes a Back-Propagation (BP) neural network to diagnose the damage state of 32 potential cracking locations in an Olympic sliding racetrack using only global settlement data from 8 strategically placed sensors. A high-fidelity finite element model generated a comprehensive training dataset, confirming that uneven settlement is the primary driver of cracking, with a stiffness degradation (SDEG) value exceeding 0.8 defined as the critical damage threshold. The trained network achieved 99 % accuracy, demonstrating its capability to provide a reliable and economically viable SHM solution. This data-driven approach significantly reduces required sensor density without compromising high diagnostic precision.

This study investigates the mechanical performance of stainless steel-timber composite shear connections equipped with stainless steel bolts. The research principally involved conducting shear push-out tests to explore the yield forms, failure modes, and the influence of parameters such as bolt diameter, spacing, and timber thickness on load-bearing capacity. The design methods for shear connections outlined in the Chinese standard GB 50005-2017, the American standard NDS-2018, and the European standard Eurocode 5 were evaluated. The results indicate that GB 50005-2017, NDS-2018 and Eurocode 5 provide conservative estimates of the bearing capacity of stainless steel-timber shear connections with Eurocode 5 offering more accurate predictions. Additionally, two analytical models, Foschi's and Hassanieh's, were evaluated and the new input parameters were proposed, with the latter model demonstrating superior accuracy in predicting the load-slip behaviour of the connections. The findings of this study contribute to the design and modeling of stainless steel-timber shear connections, providing valuable insights for future applications in structural engineering.

During the service period of steel structures, environmental corrosion has a strong impact on the mechanical and fatigue properties of steel, thereby deteriorating the structural safety. In this study, focusing on the fatigue behaviour degradation of structural steel service in the corrosive environment, the high-cycle and low-cycle fatigue properties of Q690D high-strength steel with various degrees of corrosion were investigated. The wet/ dry cyclic accelerated corrosion tests were conducted to prepare the corroded coupon specimens, and the corrosion developing characteristics were analysed. For each corrosion degree, both the Coffin-Manson relationships and the S-N curves were established, and the cyclic stress response, cyclic stress-strain curves, and cyclic energy dissipation of the corroded steel were analysed. The results indicate that with the increase in corrosion degree, the fatigue performance in both the high-cycle fatigue range and the low-cycle fatigue range undergoes an obvious deterioration. The research work in this paper discusses the fatigue performance degradation of high-strength steels after corrosion, which complements the current existing studies that focus on ordinary steels. The fatigue performance degradation across the entire life range of the high-strength steel material after corrosion has been thoroughly examined and the whole life-span Coffin-Manson relationships were established for each corrosion degree, which enabled to conduct the fatigue life assessment in both high-cycle and low-cycle fatigue regime with a unified expression. The findings can provide a valuable reference for the fatigue assessment of steel structures in corrosive service environments.

In recent years, seismic disasters have emerged as a significant threat to the safety of structures and the well-being of individuals. Stainless steel, owing to its exceptional ductility, is attracting attention for its potential use in seismic resilient buildings and infrastructures. However, current research on the seismic performance and design methods of stainless steel frames is limited. Therefore, this paper reports two cyclic loading tests on austenitic stainless steel full-scale frames with extended end-plate joints and bolt-welded joints, investigates the seismic response of components and provides experimental data for further development of seismic design methods for stainless steel structures. The cyclic loading test results for components, including the internal forces of beams and columns and the hysteresis behaviour, energy dissipation capability, deformation capacity, and stiffness of joints, were discussed in detail. On the basis of the experimental results, the applicability of existing calculation methods for the stiffness of stainless steel semi-rigid joints was evaluated. Moreover, the components of the story drift ratio of beams, columns, and joints, as well as their respective contributions to the inter-story deformation of the frame during cyclic loading were calculated and analysed. Furthermore, the sequence of plastic development in the frame during the loading process was determined through component response analyses of beams, columns and joints.