Viscoelastic damping constitutes a pivotal technology for enhancing the seismic resilience of building structures. Conventional viscoelastic dampers, typically fabricated via vulcanization, frequently exhibit insufficient bonding between steel plates and damping materials, resulting in compromised energy dissipation capacity and a susceptibility to material tearing. To overcome these limitations, this study proposes a novel Dual-phase Lattice Metamaterial Damper (DLMD) fabricated using additive manufacturing. To systematically elucidate the dynamic behavior and energy dissipation mechanisms of the DLMD, a comprehensive investigation was conducted, comprising variable-amplitude symmetric quasi-static tests, DIC, FEM analyzes and Electron Backscatter Diffraction (EBSD) analysis. The results demonstrate that the energy dissipation performance of the DLMD remains stable and exhibits negligible sensitivity to frequency within the tested range. Furthermore, under displacement amplitudes ranging from 75 % to 250 % of the design displacement (u/d), the hysteresis loops at different amplitudes are superimposable, indicating that the DLMD conforms to the Masing criterion. The DLMD exhibits a synergistic energy dissipation mode formed by the compressive/tensile deformation of the RD lattice and the bending deformation of the OCT lattice. The failure mode of DLMD involves plastic fracture of lattice truss members. The fracture zone initiates from the top RD lattice region and propagates downward, exhibiting significant deviation in its path when extending into the OCT lattice. This work offers explicit direction for the development of novel multi-phase lattice material damper featuring controllable fracture paths.

The integrity and safety of underwater bridge structures can be compromised by damage; therefore, timely detection and assessment are crucial. However, underwater damage detection is constrained by turbidity, low illumination, and multiple coexisting damage types, which complicates comprehensive automated safety assessment. This study proposes an automated framework for structural damage detection and safety assessment of underwater bridge structures. Multiple types of underwater damage are analyzed, and an underwater damage image dataset (UDID) is established. A modified linear unsharp masking method is used to adaptively enhance the high-frequency features of the UDID through multi-scale image fusion. A well-trained GoogLeNet model is used to automatically detect underwater structural damage. Based on the detection results, bridge safety is classified into five levels using the damage index method. A case study involving a concrete bridge in China demonstrates the effectiveness of the proposed framework. The GoogLeNet model achieves a detection accuracy of 97% in the large-scale test. The concrete bridge is assessed as level 3, which represents moderate damage and is consistent with field detection results. In underwater environments featured by a low signal-to-noise ratio, the detection accuracy of damage types that depend on texture and contrast features decreases significantly. This framework effectively addresses the limitations of existing underwater damage detection methods and enables automated safety assessment for underwater bridge structures.

The asynchronized parallel double-stage fluid viscous damper (APDFVD) is a novel device featuring two working stages, designed to enhance energy dissipation and damping forces under large earthquakes, thus enabling effective seismic response control across multiple earthquake intensity levels. To address the lack of a numerical model, a uniaxial APDFVD material model was developed and implemented in OpenSees. The damper's configuration, mechanism, and experimental performance are summarized, and the force calculation methods for the two working stages are introduced. The model was validated against experimental data, and the influence of key parameters was examined through parametric analysis. Subsequently, the seismic responses of a frame structure equipped with APDFVDs, conventional fluid viscous dampers (FVDs), and without dampers were comparatively analyzed under service-level earthquakes (SLE), design basis earthquakes (DBE), and maximum considered earthquakes (MCE). The results show that under SLE, the APDFVD operated in the first stage, and the control effects of the FVD and APDFVD were identical. Under DBE, although the APDFVD slightly entered the second working stage, its control effect on the maximum inter-story drift ratio (MIDR) reached 50.5 %, which was superior to that of the FVD (38.7 %), while both devices exhibited identical control effects on the maximum absolute floor acceleration (MAFA). Under MCE, the APDFVD significantly entered the second working stage, achieving reductions of 44.6 % in MIDR and 23.5 % in MAFA, both significantly greater than the corresponding values for the FVD (25.9 % and 11.9 %, respectively). These findings demonstrate that, compared to conventional FVDs, the APDFVD provides more effective control, especially under strong earthquakes, due to its displacement-dependent double-stage damping mechanism.

Isolation is widely recognized as a reliable strategy for realizing seismic resilience. Although existing research has identified the negative effects of pulse-like near-field ground motions (PNGMs) on seismically isolated structures, their specific impact on the seismic resilience of such structure remains rare, especially for the isolated shear wall structures (ISWS). Hence, a 1/10 scaled ISWS was tested using shaking table, emphasizing the structural responses differences under different types of GMs. A numerical model was then recommended and validated through comparison with the test results. The maximum relative errors between the simulated responses and seismic responses from shaking table tests were less than 5.5%. This validated model was then employed to perform numerical analysis of a real engineering practice. Under far-field ground motions (FGMs), the restoration cost and repair time amounted to 0.1% of the replacement cost and 0.5 days, respectively, corresponding to a resilience level of Level 3. Under PNGMs, these values increased to 1.96% of the replacement cost and 7.5 days, respectively, reducing the resilience level to Level 2.

Generating earthquake ground motions that match to building damage is critical for post-earthquake estimation of input ground motions based on structural damage states and for augmenting strong ground motion datasets. Existing ground motion generation methods primarily focus on simulating ground motion time histories while giving limited attention to the resulting building damage. Research on methods for generating ground motions specifically tailored to match building damage remains limited. To address this gap, this study introduces a building damage-matched ground motion generation method based on a physics-guided neural network. The proposed approach uses nonlinear time history analysis (NTHA) to compute structural seismic responses and develops a building damage classification surrogate model, referred to as the Building Damage Evaluator Model (BDEM), based on damage labels derived from these responses. By employing seismic parameters (e.g., magnitude, epicentral distance, VS30) and damage labels as conditional input variables, a conditional generative adversarial network (CGAN) is designed to generate ground motion time histories. The BDEM is embedded as a physics-based evaluator within the CGAN to guide the generation of ground motions that align with building damage states. A four-story reinforced concrete (RC) frame structure is used as a case study to validate and demonstrate the proposed method. Results indicate that the method not only accurately reproduces statistical characteristics of the waveform, spectrum, and ground motion intensity measures observed in real records but also better controls the structural damage induced by the generated ground motions. Compared to models without the embedded physics evaluator, the proportion of generated ground motions with damage labels consistent with real records increased from 55.7% to 81.5%. This method provides a valuable reference for postearthquake ground motion inversion and data augmentation in seismic damage prediction models.

Viscoelastic dampers (VEDs) provide passive and effective solutions for multi-condition structural vibration control. However, maintaining high energy dissipation across diverse temperature environments remains a critical challenge due to their temperature sensitivity. To address this, a novel temperature gradient viscoelastic wall damper (TGVWD) is developed, featuring five steel plates and four viscoelastic layers, which are strategically arranged to cover distinct optimal damping temperature intervals. This gradient configuration enables synergistic energy dissipation across a broad thermal range. Comprehensive dynamic tests at varied temperatures were conducted to assess the TGVWD's performance, with comparative analyses against a conventional viscoelastic wall damper (CVWD) highlighting its advantages. Experimental results reveal that the TGVWD exhibits superior and stable energy dissipation across a wide temperature range. Its loss factor increases initially and then decreases with rising temperature, reaching a peak at room temperature, although moderate softening occurs at elevated temperatures. Nonlinear mechanical behavior intensifies with increasing displacement, while frequency effects are relatively limited. Further comparison indicates that, under typical loading conditions, the TGVWD achieves 167.7 % and 177.0 % of the CVWD's energy dissipation per cycle at 20 degrees C and 40 degrees C, respectively, and exhibits comparable performance at lower temperatures. These results confirm the TGVWD's considerable advantages in wide-temperature applications. Furthermore, an advanced Bouc-Wen-Maxwell-Mullins (BWMM) model is developed to represent the complex hysteretic behavior by incorporating multiple nonlinear mechanisms. This study presents a robust approach to viscoelastic damping in structures subjected to fluctuating thermal environments, laying a strong foundation for both theoretical development and engineering application.

To investigate the fatigue crack growth behavior of precorroded high-strength steels, a salt spray corrosion test using compact tension specimens made of Q460C, Q550D, and Q690D is conducted, and the corrosion rate is studied by mass losses and surface roughness. The effect of precorrosion on fatigue crack growth behavior is discussed based on a fatigue crack growth test by comparing fitted material constants in the Paris law and the strain energy density factor equation with those in related literature and design codes. A prediction model to predict material constants with longer exposure time is proposed. The results show that corrosion pit depth and surface roughness generally increase with the increase in exposure time; The parameters, m and n, show a generally gradual decreasing trend while logC and logA increase and decrease, respectively, with increase in exposure time; The test data of m and logC at each exposure time demonstrate a normal distribution, and the means of the data show a linear correlation with exposure time. The proposed linear prediction model is proved feasible in predicting fatigue resistance, including m and logC in Paris law with longer exposure time, but it does not fit in predictions based on n and logA.

Objective To address the limited research on seismic response prediction for historic timber structures, the difficulty in synchronously predicting responses at multiple locations, and the insufficient validation using measured data, an intelligent multi-point seismic response prediction method was developed based on shaking-table test data. A full-scale single-eave hip-and-gable-roof timber pavilion was investigated. Deep learning models were established using column-base acceleration responses as inputs and multi-point acceleration responses of the upper structure as outputs. Long short-term memory (LSTM), gated recurrent unit (GRU), and temporal convolutional network (TCN) models were compared as baselines. Model performance was improved through convolutional-recurrent fusion, ensemble learning, and attention mechanisms. Prediction accuracy, stability, and training efficiency were systematically evaluated, and an optimal modeling strategy for multi-point seismic response prediction of historic timber structures was identified.Methods A full-scale model of a single-eave timber pavilion with a hip-and-gable roof was constructed using North China larch. Bidirectional horizontal shaking-table tests were conducted. Based on the seismic design conditions of Ying County, two recorded ground motions, RSN32 and RSN66, and an artificial ground motion (AW) matched to the code-specified target response spectrum were selected. The model was successively subjected to frequent, design-basis, and rare earthquakes at seismic intensity 8. The acceleration amplitude ratio between the primary and secondary input directions was set to 1:0.85. A multiple-input multiple-output sequence-to-sequence prediction framework was developed from the measured data. Bidirectional horizontal accelerations at the column bases were used as inputs, while those at the column heads and upper measurement points were used as outputs. LSTM, GRU, and TCN baseline models were established and enhanced using convolutional-recurrent fusion, ensemble learning, and attention mechanisms. Model hyperparameters were determined through two-fold cross-validation grouped by ground-motion type and Bayesian optimization. Data from RSN32 and RSN66 at all seismic levels were used for model training and validation, while the AW data were used as an independent test set. Prediction accuracy and generalization performance under different ground motions and seismic levels were evaluated.Results and Discussions Shaking-table tests showed that stable dynamic-response transmission was maintained by the timber pavilion under three earthquake intensity levels. The acceleration amplification factors at the measuring points were measured as 0.26-1.94 in the X direction and 0.34-2.31 in the Y direction, both lower than the typical range of 2-4 for modern building structures. Considerable energy dissipation and vibration reduction were therefore demonstrated. As the earthquake intensity increased, the acceleration amplification factors of the upper structure generally decreased, and values below 1 were recorded at several measuring points. The structural response was thus shifted from dynamic amplification under frequent earthquakes to energy dissipation and response attenuation under rare earthquakes. This behavior was attributed to sliding, rotation, and friction at the column base-stone plinth interfaces. The influence of infill-wall restraints on column-head responses was found to depend on earthquake intensity. Under frequent earthquakes, asymmetric force transmission and eccentric effects caused by unilateral restraints were more pronounced. At the design-basis intensity, local stiffness and coordinated force transmission induced by bilateral restraints were enhanced. Under rare earthquakes, the differences between unilateral and bilateral restraint points were reduced, and the structural response was governed by damage degradation and overall energy dissipation. Better overall performance and higher training efficiency were achieved by TCN than by LSTM and GRU at all earthquake intensity levels. Average coefficients of determination (R2) of 0.825 and 0.863 were obtained under frequent and design-basis earthquakes, respectively. Under rare earthquakes, the prediction accuracy of all models was reduced because of increased structural nonlinearity, local peak fluctuations, and coupling among multiple measuring points. After an attention mechanism was introduced, average mean absolute error (MAE), root mean square error (RMSE), and R2 values of 0.232 m/s2, 0.320 m/s2, and 0.676, respectively, were achieved by TCN-Attention. The highest prediction accuracy under strong earthquakes and the best cross-point stability were obtained.Conclusions The dynamic responses and intelligent prediction of historic timber structures under different seismic intensities were investigated through shake-table tests on a full-scale single-eave, hip-and-gable-roof timber pavilion and deep learning methods. Weak acceleration amplification was observed, and the amplification generally decreased with increasing seismic intensity and measurement height, indicating effective deformation coordination and seismic energy dissipation. TCN achieved higher prediction accuracy, stability, and training efficiency than LSTM and GRU under all seismic intensities and was therefore identified as an effective baseline model for predicting the seismic responses of historic timber structures. The baseline TCN provided adequate predictions under frequent and design-basis earthquakes. Under rare earthquakes, prediction accuracy and stability were substantially improved by incorporating an attention mechanism or Adaptive Boosting (AdaBoost). The best overall performance was achieved by TCN-Attention, whereas TCN-AdaBoost provided a favorable balance between accuracy and computational efficiency. Seismic responses were reduced through flexible connections, coordinated component deformation, and joint energy dissipation. The applicability of TCN and its enhanced models to strongly nonlinear seismic-response prediction was verified, providing a basis for seismic performance assessment, rapid damage prediction, structural health monitoring, and conservation and strengthening decisions for historic timber structures.

Earthquakes are severe hazards in seismic-prone areas, causing significant damage and casualties. Single-story steel industrial buildings are widely used for their high strength and durability, yet the impact of enclosure structures on their seismic vulnerability remains insufficiently studied. Taking a single-story steel industrial building in Beijing as a case study, its dynamic characteristics were obtained via operational modal analysis. An improved simplified modeling method for enclosure structures was proposed, and two finite element models (M1 without enclosure structures, M2 with enclosure structures) were established using SAP2000. Incremental dynamic analysis was conducted to evaluate seismic vulnerability with peak ground acceleration as intensity measure and maximum inter-story drift (theta max) as damage measure, and collapse margin ratio (CMR) was used to assess anti-collapse performance. Results showed the building's natural periods along X-, Y-, and torsion directions were 0.606 s, 0.427 s, and 0.431 s respectively. M2's modal period was more consistent with measured values. Under identical seismic excitation, M2 had smaller theta max and CMR values of M1 and M2 were 5.78 and 6.65, respectively. This confirms that enclosure structures reduce structural damage and enhance anti-collapse capacity. The study provides a reliable modeling method and quantitative data for the seismic performance of such buildings.

Ancient wooden architecture serves as invaluable repositories of historical and cultural heritage. Among their critical components, the dougong-fork column plays a pivotal role, yet it is susceptible to inclination or collapse resulting from material degradation and external load actions. Consequently, there is a compelling need to investigate the seismic performance of the structure. Moreover, a reinforcement method which is based on the principle of minimal intervention is proposed, involving the addition of a vertical auxiliary column positioned posterior to the existing column. To evaluate this method, full-scale models of the inclined dougong-fork column with and without the reinforcement method were constructed and subjected to pseudo-static cyclic loading tests. The analysis encompassed the damage mechanism, deformation characteristics, reinforcement mechanism, and seismic performance of the current model (CM) and the reinforced model (FM). The results indicated that vertical cracks in the fork arms were induced by the restrictive effect of the Zhutou-Fang, while local embedded pressure deformations in the Shuatou-Fang occurred due to the embedment effect of the fork column. Furthermore, under equivalent horizontal displacement, the FM exhibited an energy-dissipating capacity 1.23–3.09 times greater and a stiffness 1.67–2.13 times higher than the CM. These findings provide significant theoretical insights and practical guidance for the conservation and reinforcement of ancient wooden architecture.

A three-tower connected reinforced concrete(RC)frame building was selected as a prototype building and used to investigate resilience-based seismic design,aiming to provide a reference for multitower-connected buildings by using seismic isolation.First,a seismic resilience assessment strategy was recommended based on the characteristics of the case study.Specifically,the restoration cost index was recommended as the ratio of the total repair cost of multiple towers and connection parts with respect to the current replacement cost.In contrast,repair time and casualties were recommended as the longest repair time and the highest casualties of multiple towers due to their uncoupled functions.The influences of the critical design parameter of the isolation system(i.e.,yield ratio)on the resilient performance of the entire building was investigated.Both the repair cost and time of the building decreased at decreasing yield ratios,which were attributed to the notable control of the maximum absolute floor acceleration.Only the case study,which showed a yield ratio of 2%,achieved the highest resilience level,as regulated by the relevant Chinese code.Hence,a 2%yield ratio is recommended for the conceptual design of seismically isolated multitower-connected buildings to achieve good seismic resilience.

Structural health monitoring (SHM) is vital for ensuring the safety and durability of bridges. However, dynamic response data are typically contaminated by sensor malfunctions, environmental interference, and transmission errors, which compromise the reliability of condition assessment. Therefore, effective data cleaning is essential for SHM applications. This paper proposes a unsupervised framework that combines continuous wavelet transform with multiscale local binary pattern encoding to separate normal and abnormal data. Raw vibration signals are transformed into compact, discriminative feature vectors, thereby providing an efficient basis for clustering and automated cleaning. This method avoids the reliance on labeled data and large-scale models, thus allowing clustering to be accomplished using simple machine-learning algorithms at a low computational cost. Unlike conventional image-based deep-learning frameworks, the proposed method converts wavelet time-frequency representations into compact multiscale texture feature vectors, thereby enabling lightweight unsupervised abnormal data detection with reduced storage and computational costs. Experimental studies on bridge monitoring data demonstrate the effectiveness of the proposed framework. Compared with conventional cleaning approaches, the proposed framework achieves reliable abnormal data detection in large-scale SHM applications while maintaining high computational efficiency and scalability.

Continuous bridge monitoring generates long-duration, noisy, and multi-channel acceleration streams in which abnormal events are sparse yet safety-critical. This work targets reliable event classification under operational conditions while preserving the time-series nature of the data. A hybrid one-dimensional convolutional model with channel-wise recalibration and attention analyzes fixed-length windows from a cable-stayed bridge monitoring system, combined with peak picking, batching, and lightweight inference that suit day-scale records. Evaluation on a held-out test set shows balanced class performance: precision across classes ranges from 90.5% to 98.4%, and recall across classes ranges from 91.4% to 99.6%, yielding precision of 94.6% and recall of 94.9%. At the sensor level, classification accuracy ranges from 93.6% to 99.7%, indicating consistent behaviour across heterogeneous channels with different noise and placement conditions. The dataset is split into training, validation, and test subsets, and the test set remains untouched during model selection. The results suggest robust detection across classes and sensors, stable operation on continuous 24-h records, and practical readiness for field deployment in bridge monitoring.

The absence of both theoretical predictions regarding mechanical behavior and associated design limits has, to some extent, constrained the engineering application of annular thick rubber bearings, despite their various advantages. This work aims to develop a concise and precise approximate analytical solution to predict the mechanical behavior of annular rubber bearings under compression and to establish design limits, ultimately extending current standards to thick rubber bearings. The rigorous analytical solutions for the displacement component were derived by transforming the Navier displacement formulation into four distinct boundary-value problems. After calibration and simplification, the approximate analytical solution for displacement, strain and stress component, expressed using elementary functions and possessing clear physical interpretations, was derived. These solutions exhibit high accuracy in predicting the mechanical behavior of rubber bearings under compression, as verified by numerical simulations and existing literature. This study offers not only reliable predictive formulas for the mechanical behavior of annular rubber bearings, but also establishes relevant design limits, mainly including the max. local shear strain of rubber layer and the allowable compressive stress under service load, and the yield criteria for reinforcing steel plate, which may inform future design specifications.

The growing scarcity of freshwater and the ecological impacts of river sand mining have stimulated the search for alternative sources of mixing water and fine aggregates in the concrete industry. Seawater sea sand concrete (SWSSC) offers a promising solution by using locally available seawater and sea sand in coastal and remote island regions, thereby reducing material transportation costs and associated environmental burdens. However, the high chloride content of seawater and sea sand poses a critical challenge, namely the corrosion of ordinary steel reinforcement, which remains one of the major durability concerns for the engineering application of SWSSC. This paper systematically reviews recent progress in SWSSC, covering raw material characteristics, mix proportion design, microstructure, mechanical properties, and durability, with particular emphasis on chloride ingress and steel corrosion. Key corrosion-mitigation strategies are summarized, including stainless steel reinforcement, fiber-reinforced polymer (FRP) bars, cathodic protection systems, and high-performance coatings. The bond behavior of FRP-reinforced SWSSC composite structures, the flexural and compressive performance of beam and column members, and recent advances in FRP-tube-confined SWSSC columns are also discussed. Some representative studies have reported that, under optimized conditions such as a low water-to-binder ratio and the appropriate incorporation of slag or silica fume, the apparent chloride diffusion coefficient of SWSSC can be reduced to a very low level, suggesting a durability potential comparable to that of high-performance concrete, though further validation under long-term field exposure is needed. Practical engineering applications and existing standards are reviewed, research limitations are identified, and future research directions are proposed. Unlike previous reviews that mainly focus on material properties, durability, or individual structural types, this review integrates the complete logical chain of chloride-induced durability bottlenecks, corrosion-mitigation strategies, FRP-reinforced composite structural performance, engineering applications, and standardization gaps. The review aims to provide a systematic reference for the material design, structural application, and standard development of SWSSC.

Inter-story isolated towers built on large chassis have becoming popular in the urbanization of cities located in high-seismic regions. Shaking-table tests were conducted on 1/10-scaled two inter-story isolated towers built on a large chassis to investigate their seismic performance under near-field pulse-like ground motion (NPGM) and far-field ground motion (FGM), focusing on the effects of NPGM, FGM, and artificial ground motion. The test results indicated that the NPGM exhibited a certain extent of influence on the seismic responses of the structure, particularly under the maximum considered earthquake. The maximum isolation system displacement (MISD) and maximum inter-story drift ratio (MIDR) under the NPGM were 1.12 and 1.05 times those under the FGM. A refined numerical simulation method was recommended and validated to be capable of predicting the seismic responses of the test model. The relative errors in the MIDRs between the simulation and test results were lower than 3.1%. Next, the effects of the two types of ground motion on the prototype structure were investigated. Under the MCE, the MIDR of the upper tower and the MISD under the NPGM were 12.1% and 42.3% larger, respectively, than those under the FGM.

The Beijing courtyard house, a typical representative of traditional Chinese residential architecture, features unique micro-environmental characteristics. ENVI-met was used to simulate the microenvironment of Beijing courtyard houses, with Mao Dun’s former residence as the case study. Summer and winter scenarios were established to evaluate the key micro-environmental parameters, including Mean Radiant Temperature (MRT), air temperature, relative humidity, wind speed, and Physiological Equivalent Temperature (PET). The results indicate that MRT dominates seasonal thermal comfort in courtyard houses, correlating more strongly with PET in winter. Meanwhile, PET is also influenced by air temperature, humidity, and ventilation. In summer, hot, dry conditions intensify discomfort, while high shading rate and wind speed alleviate it. In winter, thermal comfort primarily depends on MRT during peak sunshine and air temperature during low-radiation periods. These findings reveal how thermal factors affect human thermal sensation, helping optimize courtyard thermal comfort.

This study proposes a binary connection-phase coding strategy for tailoring the mechanical response of three-dimensional re-entrant–chiral metamaterials. Two unit cells with identical global dimensions but different circumferential connection orientations are defined as a coaxial-phase state coded as 0 and an orthogonal-phase state coded as 1. Seven representative 3 × 3 × 3 configurations were fabricated from 316 L stainless steel by selective laser melting and investigated through quasi-static compression tests and finite element simulations validated against the experiments under two orthogonal loading directions. The two discrete connection-phase states share a common geometric framework, and their spatial arrangements are used to tailor collapse, apparent Poisson’s ratio, layer-wise torsion, and energy absorption. The 0-coded unit cell exhibits a long stable stress plateau and densifies at a nominal strain of approximately 0.75, whereas the 1-coded unit cell shows a higher initial stress peak, a descending plateau, and earlier densification at approximately 0.65. At the multicell scale, the all-0 configuration undergoes smoother progressive collapse, while the all-1 configuration exhibits larger stress fluctuations and more pronounced layer-wise torsion. The hybrid configurations further show that similar phase fractions do not necessarily produce similar densification and energy-absorption responses. Among the seven investigated configurations, Model E provides the highest usable SEA under Y-direction compression, Model B provides the highest usable SEA under X-direction compression, and Model G maintains effective crushing to a strain of approximately 0.64 under X-direction compression. These results demonstrate the feasibility of tailoring multiple mechanical responses through binary connection-phase coding within the sampled configuration set without changing the basic unit-cell dimensions.

Since the mechanical property prediction formulas in existing design codes are inadequate for thick rubber bearings, and the mechanical behavior of these bearings shows strong dependence on compressive stress and shear strain, this study systematically investigates their compressive and shear properties through theoretical derivation and experimental validation. First, the relevant provisions on compressive performance in representative design codes are reviewed, and the allowable compressive stress limit for rubber bearings under service loads is determined. Second, based on the Yeoh hyperelastic model, an equivalent shear modulus formulation is derived to characterize the shear-strain-dependent shear stiffness of rubber bearings. Finally, four types of specimens, including conventional and thick rubber bearings, are designed, and both compression tests and combined compression-shear tests are conducted. The experimental results indicate that, under the specified allowable compressive stress, the mechanical properties of all four types of rubber bearings remain stable and predictable. Moreover, the proposed equivalent shear modulus formulation effectively captures the nonlinear horizontal behavior characterized by initial softening followed by hardening as shear strain increases. This study aims to facilitate the engineering application of thick rubber bearings and provide a technical reference for the revision and improvement of relevant design codes.

The replaced origami-embedded honeycomb (ROHC) and the added origami-embedded honeycomb (AOHC) have attracted considerable attention due to their high energy absorption efficiency and lightweight advantages. However, their energy absorption behavior and deformation mechanisms under dynamic shear loading remain unclear. To address this gap, this study established refined finite element models to reveal the stress evolution patterns of both structures under in-plane shear velocities of 1 m/s, 20 m/s, and 50 m/s, and systematically investigated the effects of key geometric parameters of the Kresling unit on the dynamic mechanical performance of AOHC and ROHC. Research indicates that under low-velocity shear, the stress distribution in AOHC exhibits an X-shaped concentration pattern, establishing an efficient shear load transfer path through the tri-lobe plates and Kresling creases. In contrast, the stress distribution in ROHC is relatively uniform. At intermediate velocity, the stress in both structures concentrates toward the shearing end, but AOHC still retains the X-shaped characteristic. Under high-velocity shear, inertial effects dominate the deformation, leading to similar failure modes in both structures, with AOHC exhibiting better structural stability. Reasonable adjustment of the unit parameters can effectively regulate the deformation modes and specific energy absorption (SEA) of both AOHC and ROHC. AOHC outperforms ROHC in terms of SEA and peak shear force at all shear velocities. Under the same highvelocity shear and wall thickness condition (5.04 mm), the peak shear force of AOHC reaches 691.179 kN, which is approximately three times that of ROHC (228.769 kN). In summary, AOHC possesses superior shear resistance. Appropriately reducing the structural height and side length, or increasing the wall thickness, can further enhance the energy absorption capacity and stability of both structures. This study provides new insights and a reference for the design of high-performance origami honeycomb structures and their application in cushioning and energy-absorbing devices.