Frontiers of Structural and Civil Engineering最新文献

To completely solve the problem of fatigue cracking issue of orthotropic steel bridge decks (OSDs), the authors proposed a steel–ultra-high performance concrete (UHPC) lightweight composite deck (LWCD) with closed ribs in 2010. Based on the successful application of that LWCD, an adaptation incorporating an innovative composite deck structure, i.e., the hot-rolled section steel–UHPC composite deck with open ribs (SSD) is proposed in this paper, aiming to simplify the fabrication process as well as to reduce the cost of LWCD. Based on a long-span cable-stayed bridge, a design scheme is proposed and is compared with the conventional OSD scheme. Further, a finite element (FE) calculation is conducted to reflect both the global and local behavior of the SSD scheme, and it is found that the peaked stresses in the SSD components are less than the corresponding allowable values. A static test is performed for an SSD strip specimen to understand the anti-cracking behavior of the UHPC layer under negative bending moments. The static test results indicate that the UHPC layer exhibited a satisfactory tensile toughness, the UHPC tensile strength obtained from the test is 1.8 times the calculated stress by the FE model of the real bridge. In addition, the fatigue stresses of typical fatigue-prone details in the SSD are calculated and evaluated, and the influences of key design parameters on the fatigue performance of the SSD are analyzed. According to the fatigue results, the peaked stress ranges for all of the 10 fatigue-prone details are within the corresponding constant amplitude fatigue limits. Then a fatigue test is carried out for another SSD strip specimen to explore the fatigue behavior of the fillet weld between the longitudinal and transverse ribs. The specimen failed at the fillet weld after equivalent 47.5 million cycles of loading under the design fatigue stress range, indicating that the fatigue performance of the SSD could meet the fatigue design requirement. Theoretical calculations and experiments provide a basis for the promotion and application of this structure in bridge engineering.

Complex modulus (G*) is one of the important criteria for asphalt classification according to AASHTO M320-10, and is often used to predict the linear viscoelastic behavior of asphalt binders. In addition, phase angle (φ) characterizes the deformation resilience of asphalt and is used to assess the ratio between the viscous and elastic components. It is thus important to quickly and accurately estimate these two indicators. The purpose of this investigation is to construct an extreme gradient boosting (XGB) model to predict G* and φ of graphene oxide (GO) modified asphalt at medium and high temperatures. Two data sets are gathered from previously published experiments, consisting of 357 samples for G* and 339 samples for φ, and these are used to develop the XGB model using nine inputs representing the asphalt binder components. The findings show that XGB is an excellent predictor of G* and φ of GO-modified asphalt, evaluated by the coefficient of determination R² (R² = 0.990 and 0.9903 for G* and φ, respectively) and root mean square error (RMSE = 31.499 and 1.08 for G * and φ, respectively). In addition, the model’s performance is compared with experimental results and five other machine learning (ML) models to highlight its accuracy. In the final step, the Shapley additive explanations (SHAP) value analysis is conducted to assess the impact of each input and the correlation between pairs of important features on asphalt’s two physical properties.

The shear bond of interface between concrete and basalt fiber reinforced polymer (BFRP) bars during freeze-thaw (F-T) cycles is crucial for the application of BFRP bar-reinforced concrete structures in cold regions. In this study, 48 groups of pull-out specimens were designed to test the shear bond of the BFRP-concrete interface subjected to F-T cycles. The effects of concrete strength, diameter, and embedment length of BFRP rebar were investigated under numerous F-T cycles. Test results showed that a larger diameter or longer embedment length of BFRP rebar resulted in lower interfacial shear bond behavior, such as interfacial bond strength, initial stiffness, and energy absorption, after the interface goes through F-T cycles. However, higher concrete strength and fewer F-T cycles were beneficial for enhancing the interfacial bond behavior. Subsequently, a three-dimensional (3D) interfacial model based on the finite element method was developed, and the interfacial bond behavior of the specimens was analyzed in-depth. Finally, a degradation bond strength subjected to F-T cycles was predicted by a proposed mechanical model. The predictions were fully consistent with the tested results. The model demonstrated accuracy in describing the shear bond behavior of the interface under numerous F-T cycles.

The derivation and validation of analytical equations for predicting the tensile initial stiffness of thread-fixed one-side bolts (TOBs), connected to enclosed rectangular hollow section (RHS) columns, is presented in this paper. Two unknown stiffness components are considered: the TOBs connection and the enclosed RHS face. First, the trapezoidal thread of TOB, as an equivalent cantilevered beam subjected to uniformly distributed loads, is analyzed to determine the associated deformations. Based on the findings, the thread-shank serial-parallel stiffness model of TOB connection is proposed. For analysis of the tensile stiffness of the enclosed RHS face due to two bolt forces, the four sidewalls are treated as rotation constraints, thus reducing the problem to a two-dimensional plate analysis. According to the load superposition method, the deflection of the face plate is resolved into three components under various boundary and load conditions. Referring to the plate deflection theory of Timoshenko, the analytical solutions for the three deflections are derived in terms of the variables of bolt spacing, RHS thickness, height to width ratio, etc. Finally, the validity of the above stiffness equations is verified by a series of finite element (FE) models of T-stub substructures. The proposed component stiffness equations are an effective supplement to the component-based method.

Reinforced concrete (RC) flat slabs, a popular choice in construction due to their flexibility, are susceptible to sudden and brittle punching shear failure. Existing design methods often exhibit significant bias and variability. Accurate estimation of punching shear strength in RC flat slabs is crucial for effective concrete structure design and management. This study introduces a novel computation method, the jellyfish-least square support vector machine (JS-LSSVR) hybrid model, to predict punching shear strength. By combining machine learning (LSSVR) with jellyfish swarm (JS) intelligence, this hybrid model ensures precise and reliable predictions. The model’s development utilizes a real-world experimental data set. Comparison with seven established optimizers, including artificial bee colony (ABC), differential evolution (DE), genetic algorithm (GA), and others, as well as existing machine learning (ML)-based models and design codes, validates the superiority of the JS-LSSVR hybrid model. This innovative approach significantly enhances prediction accuracy, providing valuable support for civil engineers in estimating RC flat slab punching shear strength.

Deformation control constitutes one of the main technological challenges in three dimensional (3D) concrete printing, and it presents a challenge that must be addressed to achieve a precise and reliable construction process. Model-based information of the expected deformations and stresses is required to optimize the construction process in association with the specific properties of the concrete mix. In this work, a novel thermodynamically consistent finite strain constitutive model for fresh and early-age 3D-printable concrete is proposed. The model is then used to simulate the 3D concrete printing process to assess layer shapes, deformations, forces acting on substrate layers and prognoses of possible structural collapse during the layer-by-layer buildup. The constitutive formulation is based on a multiplicative split of the deformation gradient into elastic, aging and viscoplastic parts, in combination with a hyperelastic potential and considering evolving material properties to account for structural buildup or aging. One advantage of this model is the stress-update-scheme, which is similar to that of small strain plasticity and therefore enables an efficient integration with existing material routines. The constitutive model uses the particle finite element method, which serves as the simulation framework, allowing for modeling of the evolving free surfaces during the extrusion process. Computational analyses of three printed layers are used to create deformation plots, which can then be used to control the deformations during 3D concrete printing. This study offers further investigations, on the structural level, focusing on the potential structural collapse of a 3D printed concrete wall. The capability of the proposed model to simulate 3D concrete printing processes across the scales—from a few printed layers to the scale of the whole printed structure—in a unified fashion with one constitutive formulation, is demonstrated.

Training samples for deep learning networks are typically obtained through various field experiments, which require significant manpower, resource and time consumption. However, it is possible to utilize simulated data to augment the training samples. In this paper, by comparing the actual experimental model with the simulated model generated by the gprMax [1] forward simulation method, the feasibility of obtaining simulated samples through gprMax simulation is validated. Subsequently, the samples generated by gprMax forward simulation are used for training the network to detect objects in existing real samples. At the same time, aiming at the detection and intelligent recognition of road sub-surface defects, the Swin-YOLOX algorithm is introduced, and the excellence of the detection network, which is improved by augmenting the simulated samples with real samples, is further verified. By comparing the prediction performance of the object detection models, it is observed that the model trained with mixed samples achieved a recall of 94.74% and a mean average precision (mAP) of 97.71%, surpassing the model trained only on real samples by 12.95% and 15.64%, respectively. The feasibility and excellence of training the model with mixed samples are confirmed. The potential of using a fusion of simulated and existing real samples instead of repeatedly acquiring new real samples by field experiment is demonstrated by this study, thereby improving detection efficiency, saving resources, and providing a new approach to the problem of multiple interpretations in ground penetrating radar (GPR) data.

This paper proposes design charts for estimating imperative input parameters for continuum approach analysis of the nonlinear dynamic response of piles. Experimental and analytical studies using continuum approach have been conducted on single and 2 × 2 grouped piles under coupled and vertical modes of vibration, for different dynamic forces and pile depth. As these design charts are derived from model piles, the charts have been validated for prototype pile foundations using scaling law. The experimental responses of model piles are scaled up and these responses exhibit good agreement with analytical results. This study also extends to estimation of the errors in computing frequency–amplitude responses with an increase in pile length. It is found that, with an increase in pile length, the errors also increase. The effectiveness of the proposed design charts is also checked with data based on different field setups given in existing literature, and these charts are found to be valid. Thus, the developed design charts can be beneficial in estimating the input parameters for continuum approach analysis for determining the nonlinear responses of pile supported machine foundations.

Identifying crack and predicting crack propagation are critical processes for the risk assessment of engineering structures. Most traditional approaches to crack modeling are faced with issues of high computational costs and excessive computing time. To address this issue, we explore the potential of deep learning (DL) to increase the efficiency of crack detection and forecasting crack growth. However, there is no single algorithm that can fit all data sets well or can apply in all cases since specific tasks vary. In the paper, we present DL models for identifying cracks, especially on concrete surface images, and for predicting crack propagation. Firstly, SegNet and U-Net networks are used to identify concrete cracks. Stochastic gradient descent (SGD) and adaptive moment estimation (Adam) algorithms are applied to minimize loss function during iterations. Secondly, time series algorithms including gated recurrent unit (GRU) and long short-term memory (LSTM) are used to predict crack propagation. The experimental findings indicate that the U-Net is more robust and efficient than the SegNet for identifying crack segmentation and achieves the most outstanding results. For evaluation of crack propagation, GRU and LSTM are used as DL models and results show good agreement with the experimental data.

This paper introduces a machine learning approach to address the challenge of limited data resulting from costly and time-consuming fire experiments by enlarging small fire test data sets and predicting the fire resistance of reinforced concrete columns. Our approach begins by creating deep learning models, namely generative adversarial networks and variational autoencoders, to learn the spatial distribution of real fire tests. We then use these models to generate synthetic tabular samples that closely resemble realistic fire resistance values for reinforced concrete columns. The generated data are employed to train state-of-the-art machine learning techniques, including Extreme Gradient Boost, Light Gradient Boosting Machine, Categorical Boosting Algorithm, Support Vector Regression, Random Forest, Decision Tree, Multiple Linear Regression, Polynomial Regression, Support Vector Machine, Kernel Support Vector Machine, Naive Bayes, and K-Nearest Neighbors, which can predict the fire resistance of the columns through regression and classification. Machine learning analyses achieved highly accurate predictions of fire resistance values, outperforming traditional models that relied solely on limited experimental data. Our study highlights the potential for using machine learning and deep learning analyses to revolutionize the field of structural engineering by improving the accuracy and efficiency of fire resistance evaluations while reducing the reliance on costly and time-consuming experiments.