Cracks and bugholes (surface air voids) are common factors that affect the quality of concrete surfaces, so it is necessary to detect them on concrete surfaces. To improve the accuracy and efficiency of the detection, this research implements a novel deep learning technique based on DeepLabv3+ to detect cracks and bugholes on concrete surfaces. Firstly, in the decoder, the 3×3 convolution of the feature fusion part is improved to a 3-layer depth separable convolution to reduce the information loss during up sampling. Secondly, the original expansion rate combination is changed from 1, 6, 12, 18 to 1, 2, 4, 8 to improve the segmentation effect of the model on the image. Thirdly, a weight value is added to each channel of the Atrous Spatial Pyramid Polling (ASSP) module, and the feature maps that contribute significantly to the target prediction are learned and screened. To use this method, a database is built containing 16, 662 256×256 pixel images of bugholes and cracks on concrete surfaces. The two defects included in those images are labeled manually. The DeepLabv3+ architecture is then modified, trained, validated and tested using this database. A strategy of model-based transfer learning is applied to optimize and accelerate the learning efficiency of the model. The weights and biases of the Xception part of the model are initialized by the pretrained backbones. The results are 97.63% (crack), 93.53% (bughole) Average Precision (AP), 95.58% Mean Average Precision (MAP) and 81.87% Mean Intersection over Union (MIoU). A comparative study is conducted to verify the performance of the proposed method, and the results demonstrate that the proposed approach performs significantly better in crack and bughole detection on concrete surfaces.

Construction site layout planning (CSLP) that considers multi-objective optimization problems is essential to achieving sustainable construction. Previous CSLP optimization methods have applied to traditional cast-in-place buildings, and they lack the application for sustainable prefabricated buildings. Furthermore, commonly used heuristic algorithms still have room for improvement regarding the search range and computational efficiency of optimal solution acquisition. Therefore, this study proposes an improved multi-objective optimization and decision-making method for layout planning on the construction sites of prefabricated buildings (CSPB). Firstly, the construction site and temporary facilities are expressed mathematically. Then, relevant constraints are determined according to the principles of CSLP. Ten factors affecting the layout planning on the CSPB are identified and incorporated into the method of layout planning on the CSPB in different ways. Based on the elitist non-dominated sorting genetic algorithm (NSGA-II), an improved multiple population constraint NSGA-II (MPC-NSGA-II) is proposed. This introduces the multi-population strategy and immigration operator to expand the search range of the algorithm and improve its computational efficiency. Combined with the entropy weight and technique for order preference by similarity to an ideal solution (TOPSIS), improved multi-objective optimization and decision for the CSLP model is developed on the CSPB. Practical cases verify the effectiveness and superiority of the algorithm and model. It is found that the proposed MPC-NSGA-II can solve the drawbacks of the premature and low computational efficiency of NSGA-II for multi-constrained and multi-objective optimization problems. In the layout planning on the CSPB, the MPC-NSGA-II algorithm can improve the quality of the optimal solution and reduce the solution time by 75%.

Many structures in civil engineering are symmetrical. Crack detection is a critical task in the monitoring and inspection of civil engineering structures. This study implements a lightweight neural network based on the YOLOv4 algorithm to detect concrete surface cracks. In the extraction of backbone and the design of neck and head, the symmetry concept is adopted. The model modules are improved to reduce the depth and complexity of the overall network structure. Meanwhile, the separable convolution is used to realize spatial convolution, and the SPP and PANet modules are improved to reduce the model parameters. The convolutional layer and batch normalization layer are merged to improve the model inference speed. In addition, using the focal loss function for reference, the loss function of object detection network is improved to balance the proportion of the cracks and the background samples. To comprehensively evaluate the performance of the improved method, 10,000 images (256 × 256 pixels in size) of cracks on concrete surfaces are collected to build the database. The improved YOLOv4 model achieves an mAP of 94.09% with 8.04 M and 0.64 GMacs. The results show that the improved model is satisfactory in mAP, and the model size and calculation amount are greatly reduced. This performs better in terms of real-time detection on concrete surface cracks.

Long-span highway–railway double-deck truss bridges are mostly located in urban centers, where noise pollution and traffic safety issues have a great impact. The fully enclosed barrier has excellent sound insulation and windproof performance and has been widely used in double-deck truss bridges in recent years. However, the large volume and the low air permeability rate will affect the aerodynamic characteristics of the bridge, resulting in vortex-induced vibration (VIV). To analyze how the fully enclosed barrier influences the highway–railway bridge VIV performance, this study analyzes the Huangjuetuo Yangtze River Bridge, combined with the wind tunnel test and the numerical calculation method to study different variations of the static three-force coefficient, the change of VIV response, and its influence mechanism after setting the fully enclosed barrier. The results show that the static three-force coefficient of the double-deck truss bridge changes significantly, the drag coefficient increases, and the absolute values of the lift coefficient and the moment coefficient decrease after the fully enclosed barrier is set. The aerodynamic performance of the bridge is significantly reduced after the fully enclosed barrier is set, and the amplitude and range of the VIV response are increased. Vertical bending VIV increased by an average of 58.5%, and torsional VIV increased by an average of 21.9%. Considering driving comfort and safety, attention should be paid to the double-deck truss bridge with a fully enclosed barrier.

The long-span double-deck truss girder bridge has become a recommend structural form because of its good performance on traffic capacity. However, the vortex-induced vibration (VIV) characteristics for double-deck truss girders are more complicated and there is a lack of related research. In this research, wind tunnel tests were utilized to investigate the VIV characteristics of a large-span double-deck truss girder bridge. Meanwhile, the VIV suppression effect of the aerodynamic mitigation measures was measured. Furthermore, the VIV suppression mechanism was studied from the perspective of vortex shedding characteristics. The results indicated that the double-deck truss girder had a significant VIV when the wind attack angles were +3° and +5°. The aerodynamic mitigation measures had an influence on the VIV response of the double-deck truss girder. The upper chord fairing and lower chord inverted L-shaped deflector plate played a crucial role in suppressing VIV. Numerical analysis indicated that vortex shedding above the upper deck or in the wake region may dominate vertical VIV, while vortex shedding in the wake region of the lower deck may dominate torsional VIV. The upper chord fairing and lower chord inverted L-shaped deflector plate disrupted the original vortex shedding pattern in both regions, thereby suppressing VIV. This research can provide a foundation for bridge design and vibration suppression measures for large-span double-deck truss girder bridges.

Fire safety evacuation has been used in numerous different kinds of buildings. This research conducts a scientometric review of fire safety evacuation applications and advances in the buildings to clarify the research trends of fire evacuation in the future and provide guidance for relevant research. A total of 3312 journals and conference proceedings were analyzed through different dimensions. The result proves that evacuation environments concentrate mainly on residential building, commercial building, school, and railway station. The characteristics of the evacuee have been gradually refined in recent years, including children, the elderly, patients, and vulnerable groups. The main experimental approaches of fire safety evacuation are evacuation drills, site records, and VR/AR experiments. The crowd behavior models mainly consist of six types: a cellular automata model, a social force model, a lattice gas model, a game-theoretic model, an animal agent-based model, and a computer agent-based model. The analysis results in the theoretical method are becoming gradually closer to the behavioral characteristics and movement data of the crowd during the actual evacuation with improvements of practical considerations. The study of evacuation drills, disaster rescue, emergencies, and other external environmental factors will become the forefront of future research, and subway stations, airports, high-rise building, and other personnel places will be the focus of the study of crowd evacuation.