本文将卷积神经网络与DIC（Digital Image Correlation， DIC）技术相结合，解决了装配式钢筋混凝土（Reinforced Concrete，RC）梁内部损伤识别问题。提出了一种用于检测RC梁内部损伤的卷积神经网络模型，并基于ABAQUS二次开发功能制作了损伤RC梁的位移云图数据集，将位移云图作为模型输入，利用卷积神经网络优异的图像处理能力来反演结构内部的损伤指标，并提出了一种三维交并比网络优化算法提高网络性能。通过DIC技术捕捉试验室浇筑的混凝土损伤构件的位移云图，对该方法进行测试与验证，实现了对结构内部的损伤识别。研究结果表明，该损伤识别方法可实现RC梁内部较高精度与鲁棒性的损伤定位。为预制装配式构件出厂质量检测提供了新思路。

该文将卷积神经网络与DIC(Digital Image Correlation, DIC)技术相结合，解决了装配式钢筋混凝土(Reinforced Concrete，RC)梁内部损伤识别问题。提出了一种用于检测RC梁内部损伤的卷积神经网络模型，并基于ABAQUS二次开发功能制作了损伤RC梁的位移云图数据集，将位移云图作为模型输入，利用卷积神经网络优异的图像处理能力来反演结构内部的损伤指标，并提出了一种三维交并比网络优化算法提高网络性能。通过DIC技术捕捉试验室浇筑的混凝土损伤构件的位移云图，对该方法进行测试与验证，实现了对结构内部的损伤识别。研究结果表明，该损伤识别方法可实现RC梁内部较...

Cyclic temperatures can induce significant structural effects on long-span cable-stayed bridges, challenging their safe operation. However, the thermal analysis of real bridges throughout their service life is computationally consuming, especially for long-span bridges. Therefore, it is essential to improve the computational efficiency for long-term structural temperature field simulation. This study proposes a simple-to-use data-driven approach for modeling the stochastic temperature field using monitoring data while considering time-varying temperature characteristics. Its key concept is that the temperature variances can be decomposed into periodic daily fluctuations and time-varying daily temperature extremes. The daily temperature extremes are modeled using Gaussian process regression and Copula theory, while their occurrence times are determined from the measured data using kernel density estimation. The approach is validated by comparing simulated and measured temperatures, focusing on both time histories and probability distributions. The results show a good agreement between the simulated and measured temperatures, confirming the validity of the proposed approach. © 2025

In recent years, demands for mitigating structural damage and post-earthquake repair costs have spurred considerable research on self-centering seismic force-resisting systems. This study evaluates the Post-tensioned Precast concrete Frame with non-grouted fuse-type Dissipaters (PPFD) through experimental and numerical methods. Quasi-static tests were conducted on a one-bay, single-story PPFD to investigate its seismic response. The tested PPFD was shown to effectively control damage to concrete components and limit residual displacements; imperfections at the beam-column interface joint had a negative effect on the initial stiffness and self-centering performance of the PPFD. Parametric analyses were then performed to discuss the force and deformation responses of the beam-column joint in the PPFD with dissipaters solely beneath the beam. A design method for the PPFD was developed based on the formulas derived and verified from the parametric analyses. Nonlinear dynamic analyses were then employed to evaluate prototype designed multi-story PPFDs subjected to individual earthquakes in addition to mainshock-aftershock sequences. The findings indicated that both configurations of PPFD, with dissipaters either symmetrically arranged on the beam or located exclusively below the beam, exhibited comparable maximum inter-story drifts and significantly smaller residual inter-story drifts than a comparable reinforced concrete frame. Moreover, the PPFD demonstrated only a minor increase in both the maximum and residual inter-story drifts when subjected to aftershocks, compared to a much larger increase for the reinforced concrete frame. © 2024 Elsevier Ltd

In this study, seven emulative precast concrete (PC) beam-column subassemblies with or without connecting bars and one concrete (RC) were fabricated and tested to investigate the effect of various types of anchorage detailing for reinforcing and connecting bars on the progressive collapse resistances. Two anchorage detailing formats for steel bars with bending and straight ends were used in the PC specimens for both the reinforcing and connecting bars. Test results show that, as found in the RC specimen, resisting mechanisms including pure flexure, compressive arch action (CAA) and tensile catenary action (TCA) were sequentially activated in all PC specimens. It is demonstrated that the connecting bars improved the progressive collapse resistance in both the CAA and TCA stages. The effect of the anchorage detailing was found to be negligible in CAA stage. However, in the TCA stage, the proper usage of the straight end for steel bars was helpful to improve the deformation capacity and thereafter enhanced the progressive collapse resistance. When the loading resistance, the deformation capacity as well as the construction flexibility are taken into account, the configuration consisting of both reinforcing and connecting bars with straight end as anchorage detailing is recommended for the design of precast beam-column connections in case a central column removal scenario is assumed. © 2024 Elsevier Ltd

Bundled reinforcement can improve construction efficiency by reducing the number of connectors required. The steel -fiber reinforced polymer (FRP) composite bar (SFCB) is a novel reinforcement with high strength, moderate modulus of elasticity and good durability. Four groups of pullout tests were carried out, with parameters including the number of bars within each bundle, bonded length, and rebar type (steel bar, BFRP bar, SFCB). The failure modes of bar pullout without or after yielding, rebar fracture, and concrete splitting were observed. Compared to the single -bar specimen, the two -bar and three -bar bundled specimens resulted in approximately 30% and 40% decrease in average bond strength. The yielding slip sy of the three -bar bundled SFCB specimen is 3.8 times that of the single -bar specimen, indicating that bundled reinforcement can effectively control sy. A simplified bond strength prediction method and a bond stress -slip constitutive model based on the equivalent effective area approach are proposed. Finite element (FE) models were established and verified, and the failure modes of bar fracture are obtained by varying the bonded lengths for different bundled reinforcements, the ultimate slip of four -bar bundled BFRP bars increased by 127%. Recommendations for the development length of bundled SFCBs are proposed based on the FE results, which provide enough safety redundancy.

The robustness of precast concrete frame structures to withstand extreme events like explosions, impacts, fires and terrorist attacks has attracted growing attention. With the understanding of the static progressive collapse performances of 2D precast frame structures, research on the dynamic responses of 3D precast frame structures is pressing and high in demand. To evaluate the influence of floor slabs on dynamic progressive collapse responses of 3D precast concrete frame structures, an experimental program consisting of a beam-column (PC) and a beam-slab-column (PCS) specimens was first designed, followed by the analysis of the working mechanisms to resist progressive collapse. The modeling techniques for 2D PC and PCS substructures were then proposed and validated using the test results. The numerical strategy was further extended to simulate the dynamic progressive collapse responses of 3D PC- and PCS-frame structures. Results show that, under various column loss scenarios, floor slabs significantly improve progressive collapse resistances of 3D frame structures. Moreover, it reveals that the floor slabs have a remarkable effect on reducing the nodal displacements and the dynamic amplification factors. © 2024 Institution of Structural Engineers

With the continuous growth of traffic load, maintenance has become the main task of various management departments. The cable force loss is directly related to the long-term service performance and life of the bridge. The wireless electro-mechanical impedance (EMI) technique based on PZT transducer attracts more and more attention due to its simple operation, fast detection speed, and wide application range. In this paper, the PZT transducers were coupled to the surface of the nut of the cable to monitor the cable force. Then the influence of the number of EMI sampling points (i.e., 250, 500, 1000) on the RMSD was investigated. The feasibility of the wired and wireless EMI method applied on the cable force monitoring was verified. The results show that the peak of the impedance curve shifted to the right with the increase of the force. The sampling points had little influence on the normalized RMSD. Therefore, the smaller sampling points could be selected to reduce power consumption. The PZT transducer has a strong potential to have a uniform sensitivity coefficient for a member under test. © Tongji University Press 2024.

In seismology and earthquake engineering, it is fundamental to identify and characterize the pulse-like features in pulse-type ground motions. To capture the pulses that dominate structural responses, this study establishes congruence and shift relationships between response spectrum surfaces. A similarity search between spectrum surfaces, supplemented with a similarity search in time series, has been applied to characterize the pulse-like features in pulse-type ground motions. The identified pulses are tested in predicting the rocking consequences of slender rectangular blocks under the original ground motions. Generally, the prediction is promising for the majority of the ground motions where the dominant pulse is correctly identified. © 2024, Institute of Engineering Mechanics, China Earthquake Administration.

Machine learning-based methods, especially deep learning methods, have achieved great success in seismic response modeling due to their exceptional performance in capturing nonlinear features. However, imbalanced features of a limited training data set can significantly decrease the prediction accuracy of machine learning models. Therefore, this study proposes a novel frequency-based clustering approach for ground motion selection to generate a balanced training data set to improve the data-driven surrogate modeling of bridges. The hierarchical clustering method was developed to suppress the redundant information on the basis of a wavelet analysis of ground motion records. The proposed method was validated by a benchmark finite-element model of a girder bridge, in which long short-term memory (LSTM) neural network was used to predict the seismic responses given ground motion excitations. Specifically, the prediction performances of LSTM surrogate models trained using different data sets have been compared, while the influence of time-frequency characteristics of ground motions has been discussed in detail. The results indicated that the proposed method can provide a balanced training data set with a uniform distribution of time-frequency characteristics and effectively improve the prediction accuracy of deep learning-based surrogate models. © 2023 American Society of Civil Engineers.