Traditional bridge inspection methods, such as visual inspections and basic non-destructive tests, remain foundational but face limitations due to labor intensity and dependency on inspector expertise, often resulting in inconsistencies. Machine learning (ML) tools offer transformative solutions by enabling the analysis of large datasets to enhance damage detection accuracy, optimize maintenance schedules, and improve resource allocation. The objective of this study was to explore the integration of advanced inspection tools, such as Unmanned Aerial Vehicles (UAVs), digital imaging, and fiber optic sensors, with ML models to enhance data accuracy, decision-making, and inspection rating efficiency. A state-of-the-art review is conducted in this study on the application of ML techniques in bridge inspection and maintenance, encompassing 60 articles and reports published in the last decade. The results of this study show that ML’s ability to integrate advanced inspection tools to improve data accuracy, decision-making, and inspection efficiency. However, the existing challenges persist in data quality, model generalization, and the need for standardization approaches across diverse bridge conditions. The review provides insights into current methodologies, benefits, limitations, and future directions, emphasizing ML’s pivotal role in modernizing bridge health monitoring for a safer and more sustainable infrastructure network. The findings of this study are expected to assist transportation agencies in planning and operating predictive maintenance strategies, focusing on damage detection, maintenance optimization, and resource allocation.

As the aging of concrete structures becomes increasingly severe, crack detection has become a crucial aspect of maintaining their sustainable use. This paper investigates an automatic detection and analysis method for concrete cracks based on deep convolutional neural network architecture. The study utilized drones equipped with high-definition cameras to capture images of concrete structures such as bridges and roads, collecting over 10,000 original images containing cracks. Using Labelme software for point-by-point image annotation and applying data augmentation techniques like rotation and scaling, the dataset was expanded to 30,000 images to meet the requirements of deep learning model training. An improved AlexNet model was developed, replacing the fully connected layer with global average pooling to enhance robustness and reduce overfitting. The model achieved an average accuracy of 95.42% on the test set, outperforming traditional methods by 15%. The model also proved effective in noisy environments. Additionally, a semantic segmentation model based on Fully Convolutional Network (FCN) was introduced, incorporating atrous convolution and spatial pyramid pooling, achieving a pixel-level accuracy of 97.6%, surpassing benchmark models such as FCN and U-Net. The model accurately estimated crack width, with an error rate within ±2% compared to field measurements. This method improves detection efficiency and accuracy while reducing manual intervention, providing strong support for the maintenance of concrete structures.

Accurate estimation of road roughness is crucial for vehicle dynamics analysis and road performance evaluation. To enhance the accuracy of estimating road roughness, this study introduces an estimation method utilizing the improved Kalman filter (KF) with discrete trapezoidal load. Firstly, the theoretical formulation of the vehicle response under road roughness load is derived, and the system equation is derived in a continuous time state-space form. Next, the vehicle system equation is discretized and an improved KF algorithm with trapezoidal load is proposed, along with the derivation of the structural state estimation formula. Then, a vehicle model with seven degrees of freedom (DOFs) is established for numerical calculation, with the improved KF algorithm is utilized to estimate road roughness based on the vehicle response. Finally, the estimation accuracy of the proposed method is verified via field tests involving the vehicle driving through bumps and driving on a rough road.

GFRP rebar is a desirable alternative to steel rebar especially in harsh environments with advantages of light weight, high strength, stable chemical properties, and corrosion resistance. However, there is a lack of understanding on the durability of GFRP rebar which significantly limits its engineering applications. Unfortunately, it is almost impossible to develop any simple theoretical model to predict the residual strength of GFRP rebar after environmental exposure. To effectively address this research gap, this paper investigates machine learning models to predict the residual tensile strength of GFRP rebar due to environmental degradation. Firstly, a database was built from the literature containing a total of 350 tensile testing results of GFRP rebar after experimental exposure. Six key influencing parameters were considered, including fiber content, bar diameter, resin type, exposure temperature, pH, and aging time. Secondly, machine learning models were trained and tested using the database, and the selected models included Decision Tree, Random Forest, Support Vector Machine (SVM), Multilayer Perceptron (MLP), and Long and Short-Term Memory (LSTM) models. The LSTM model demonstrated the best performance in strength prediction, achieving an R2 of 0.96 on the training set and 0.91 on the testing set. Thirdly, the influencing parameters were ranked using SHAP and Random Forest in terms of their impact on the residual tensile strength of GFRP rebar, and it was found that temperature has the most significant effect followed by fiber content, exposure time, pH and bar diameter, while resin type showed the least importance. Notably, SHAP analysis also showed that the coupling of different parameters also had impact on the residual strength, and the combined effect of fiber content with other parameters was the most prominent. This paper demonstrates the feasibility of machine learning models in the durability study of GFRP composite materials.

Bridge influence line has an important application in structural performance evaluation, damage identification, model correction and bridge weigh-in-motion. Influence line identification method based on dynamic testing is often interfered by the vehicle-bridge coupling vibration and the measurement noise. A deflection influence line identification method based on VMD and B-spline basis function was proposed in this article. The VMD was introduced to eliminate the dynamic fluctuation component, and then the quasi-static response could be extracted. The optimal decomposition parameters of VMD were determined by permutation entropy and GWO algorithm. The B-spline basis function was used to expand the influence line, and the identification of influence line was transformed to solve the weight coefficients of the basis function, which would greatly reduce the number of unknown variables and smooth the local fluctuations caused by noise. Finally, the feasibility and accuracy of the proposed method were verified by a numerical simulating and experimental testing, and the influence of various parameters such as vehicle speed, vehicle weight and road roughness were investigated. The results showed that the proposed method could effectively eliminate the fluctuation disturbance, and the deflection influence line could be identified with accuracy acceptable for engineering applications.

Based on the excellent mechanical properties of ultra high performance concrete (UHPC) such as high strength and good toughness, this study proposed an ultra high performance concrete-normal concrete (UHPC-NC) hybrid beams, in which UHPC was used to replace part of the NC in the tension or compression zone along the beams. The flexural behavior of 4 UHPC-NC hybrid beams and 2 control beams at different UHPC height was investigated. Based on the flexural experiments of UHPC-NC hybrid beam, a finite element model was established using ABAQUS to investigate the effects of UHPC height, reinforcement ratio, NC compressive strength and UHPC tensile strength on the flexural behavior of the hybrid beams. In addition, the differences in flexural performance and force mechanisms between the hybrid beams and the control beams were compared. Finally, based on the immediate stiffness and effective cross-sectional moment of inertia of the hybrid beams, a model for the mid-span deflection calculation of the UHPC-NC hybrid beam at a reasonable UHPC height was proposed. The results show that the load carrying capacity, energy absorption capacity and stiffness of the hybrid beams increase with the UHPC height, while the ductility initially increases and then decreases. The UHPC height has a significant effect on the flexural performance of UHPC-NC hybrid beams. The reinforcement ratio has positively improved the load carrying capacity of the hybrid beams, although specimens with larger reinforcement ratios exhibit lower ductility. The mid-span deflection calculation model for UHPC-NC hybrid beams was proposed in this paper is in good agreement with the existing experimental results, the calculation process of the calculation model is relatively simple and can provide a theoretical basis for engineering design.

Suspension bridges’ excellent spanning capacity is compromised by their insufficient stiffness, making their live load response quite challenging. This study introduces an analytical method for solving the response of the suspension bridge under the temperature-vehicle combined loading. The deformation of the main cable is estimated based on the elastic catenary theory. The variation of internal forces expresses the pylon’s and the beam’s deformation. Displacements of key nodes are estimated and set as constraints. Given the high non-linearity and complexity of suspension bridges, the residual sum of squares of all constraints is the objective function in an optimization algorithm. The analytical solutions to the system of equations are found by employing the generalized reduced gradient method (GRG), yielding deformations and internal forces of the suspension bridge. The proposed method demonstrates high computational efficiency and accuracy, achieving a computation time of less than 1 min for a 1700 m suspension bridge and a maximum vertical displacement error of only 0.59%. Further analysis reveals that the coupling effect between temperature and vehicle loads has a limited impact on vertical displacement calculations but causes a non-negligible influence on beam-end rotations. These findings highlight the necessity of considering temperature-vehicle interaction in suspension bridge assessment to ensure more precise and reliable predictions.

Laser scanning (LS) is a powerful tool for acquiring geometric information about structures. However, they face challenges in capturing targets and computing edges. This paper presents a method for automatically identifying point cloud (PC) planes and obtaining geometric information, which can provide construction quality. The proposed method consists of two parts. First, an iterative growth algorithm is introduced to identify multiple planes simultaneously with fewer parameters, which improves upon the traditional plane detection method that needs better robustness due to the abundance of parameters. Second, a framework is introduced for computing geometric information of PC planes, which integrates a k-dimensional tree to replace local density in the density peak clustering algorithm and computes edges while considering mixed points, significantly reducing the time and increasing precision. The effectiveness and accuracy of the method are verified by structural geometric information in the construction process.

This study experimentally and analytically investigates the ultimate axial strengths of carbon fibre reinforced polymer (CFRP) confined circularized rectangular concrete columns under monotonic and cyclic compression. 12 concrete columns were divided into four groups. The first group was a control group. Columns of the second, third, and fourth groups were circularized to a diameter of 267 mm and then wrapped with low CFRP confinement configurations F1, F2, and F3 of 0.33, 0.67, and 1.0 layers, respectively. Of each group, the first column was monotonically loaded, while the other two columns were subjected to cyclic loadings. The results indicated that the control specimens experienced brittle failure with low ultimate lateral strains and inclined failure surfaces. In contrast, both circularization and CFRP wrapping provide a favourable condition to significantly increase the lateral strain capacity up to 300 times that of the control columns. For partly wrapped columns of groups F1 and F2, concrete failure initiated at the unwrapped region, which caused local stress concentration on CFRP, leading to failure of CFRP and then failure of columns. For fully wrapped specimens of group F3, the failure of columns occurred at a point and then progressively developed to a larger area of CFRP. Compared with control columns, the combination of circularization and CFRP retrofitting increased the ultimate strengths of concrete column groups F1, F2, and F3 by 267%, 304%, and 394%, and the ultimate lateral strain by 744.2, 628.5, and 367.7 times, respectively. This is attributed to the mutually beneficial interaction among the CFRP wrapping, circularized concrete, and original concrete. Due to its mitigation of the local failure, cyclic loading slightly increased the axial strength, while it decreased the ultimate lateral strain of CFRP confined circularized concrete columns compared with monotonic loading. The available Lam and Teng model reasonably predicts the ultimate axial strength of CFRP confined circularized concrete columns.

This paper presents the development of stress-strain models for the confined concrete in square fibre-reinforced polymer (FRP)-concrete-steel hybrid multitube concrete columns (MTCCs) subjected to uniaxial compression. A square MTCC consists of a square FRP outer tube and multiple inner steel tubes, with concrete filling the spaces inside all tubes. In comparison to traditional square concrete columns, square MTCCs have a number of advantages, such as ample ductility and excellent durability. The compressive behaviour of square MTCCs has been experimentally investigated to demonstrate its structural advantages in the existing studies. The experimental results confirmed that the concrete in square MTCCs is well confined despite its square cross-section. However, the complex confinement mechanism of square MTCCs is challenging to be completely understood experimentally due to the difficulties associated with the measurement of the nonuniform distributions of stresses across the cross-section. With the employment of the finite element (FE) method, the complex stress distribution and the interactions between the three components can be captured to fully explore the confinement mechanism of square MTCCs. In this paper, the development of three-dimensional FE models for square MTCCs is first presented. A parametric study using the validated FE models is then presented, which was used to generate a database for the establishment of stress-strain models for the confined concrete in MTCCs.