Yanran Shi

Building façade damage segmentation remains challenging because cracks, peeling, hollow-like areas, stains, and erosion often exhibit weak contrast, irregular boundaries, discontinuous structures, and substantial scale variations under complex surface textures and illumination conditions. To address these challenges, this study proposes MMDF-Net, a multimodal instance segmentation network for paired RGB–IR façade damage images. The network adopts dual RGB and infrared branches and performs multi-scale cross-modal interaction across P2–P5 levels. A Frequency-Decoupled Cross-Modal Bridge is designed to separately model low-frequency material responses and high-frequency local damage details, thereby enabling gated bidirectional exchange between the two modalities. A Crack Topology-Aware Encoder further enhances the continuity of thin, curved, branched, and locally discontinuous cracks through directional strip convolution and bending-aware modeling. In addition, a Multi-Scale Damage Decoding Pyramid integrates high-resolution boundary details, mid-level structural cues, and high-level semantic context to support stable mask prediction across different damage scales. Experiments were conducted on a custom-built RGB–IR building façade damage dataset containing 1500 paired image samples across five damage categories. MMDF-Net achieved 82.1% mask precision, 78.6% mask recall, 72.3% mask mAP50, and 65.6% mask mAP50–95, with 7.8 M parameters and 23.5 GFLOPs, outperforming representative RGB and RGB–IR segmentation baselines. These results indicate that task-oriented multimodal feature organization can improve façade damage segmentation without relying on excessive model expansion.

Wide-area visual monitoring of construction sites is constrained by the reliable detection of safety-critical targets that appear small, low-resolution, and weakly textured under elevated or distant camera views. To address this problem, this study proposes Scale-Gated Edge Detection Transformer (SGE-DETR), a safety-oriented end-to-end detector for large open construction scenes. The framework incorporates scale-aware residual edge modulation to preserve weak contours and local structures, density-guided context-adaptive fusion to balance multi-level features according to contextual and edge-density responses, and spatial gated reparameterized feature refinement to suppress redundant background textures. Experiments were conducted on SODA and STWD using COCO-style scale-sensitive metrics and efficiency indicators. On SODA, SGE-DETR achieved AP50, APS, APM, and APL values of 0.8748, 0.2157, 0.4577, and 0.6013, respectively, with 32.5 GFLOPs, 14.5 M parameters, and 83.4 FPS. On STWD, it obtained the highest AP50, APS, APM, and APL among the compared methods, reaching 0.7936, 0.8132, 0.8594, and 0.9253, respectively. Ablation results further showed that the full model improved mAP50 and mAP50–95 over RT-DETR-r18 by 4.15 and 2.93 percentage points while reducing computational complexity. These results indicate that SGE-DETR improves safety-oriented small-object detection and multi-scale robustness while retaining a relatively low parameter count.

To reveal the dynamic evolution mechanism of hydraulic fracturing in hydraulic tunnels under different confining pressures and overcome the limitations of traditional tests and numerical methods in simulating complex crack propagation and multi-field coupling problems, a meshless numerical simulation framework suitable for hydraulic fracturing in hydraulic tunnels is constructed using the Smoothed Particle Hydrodynamics (SPH) method. Through the definition of the contact force transmission mechanism between matrix particles and water particles as well as the particle failure criterion, the dynamic tracking of the entire process of crack initiation, propagation, and penetration is realized. Multiple schemes with confining pressure ratios λ (σ_x/σ_y) of 0.2, 0.4, 0.6, and 0.8 are set up to systematically investigate the influence of confining pressure ratio on the hydraulic fracturing process. The research results show that the SPH method can effectively reproduce the evolution of complex crack networks in hydraulic fracturing and overcome the defects of traditional grid methods in dealing with discontinuity problems. The confining pressure ratio is a key parameter regulating the crack morphology. With the increase of λ, the crack network gradually transforms from a dendritic shape (λ = 0.2) to a m-shaped pattern (λ = 0.4) and a snowflake-like structure (λ = 0.6, 0.8). The increase in the proportion of horizontal stress significantly promotes the lateral propagation of secondary cracks. A common mechanism of “corner stress concentration dominating crack initiation” exists under different confining pressure ratios. Moreover, with the simultaneous increase of σ_x and σ_y, the degree of stress concentration decreases, and the crack propagation rate slows down. This study provides theoretical support and quantitative tools for the analysis of water-mechanics coupling disaster-causing mechanisms and engineering disaster prevention and control in hydraulic tunnels.

Polypropylene fibers (PPFs), characterized by their low density, cost-effectiveness, and superior corrosion resistance, can be effectively incorporated into concrete to enhance the impact resistance of wall panels. This study introduces an innovative composite wall panel utilizing polypropylene fiber-reinforced concrete (PFRC) as the core material. Initially, an experimental investigation into the mechanical properties of PFRC was conducted, and based on these results, a constitutive model for PFRC was established. Subsequently, the impact-induced mechanical behavior of the innovative composite wall panel was investigated through finite element simulations employing ABAQUS, version 2020, software. The findings indicate that polypropylene fibers significantly improve both the compressive strength and ductility of concrete, with an optimal coarse fiber content of 1%. The inclusion of glass fiber grids and polypropylene fibers reduced the number of cracks and the overall deformation of the composite wall panel. The integration of glass fiber grids coupled with fiber reinforcement resulted in 7.2% and 27.8% enhancements in impact resistance, respectively. Parametric studies demonstrated that greater concrete panel thickness effectively diminishes post-impact peak and residual displacements in composite wall systems. Furthermore, the impact resistance was found to be weaker at the panel edges and stronger at a quarter of the panel height.

Hydraulic concrete is prone to cracking and interactive propagation under complex stress, threatening its structural integrity and service life. To address limitations of traditional numerical methods (e.g., mesh dependency in FEM) and imprecision of existing meshless methods for characterizing multi-fissure interactions, this study improved SPH to model double-crack interactions in hydraulic concrete under three-point bending and clarify the underlying mechanisms. A modified SPH framework was developed by introducing a failure parameter (ξ) to refine the kernel function, enabling simulation of particle progressive failure via the Mohr–Coulomb criterion; a three-point bending numerical model of concrete beams containing double precast fissures (induced and obstacle) was established, with simulations under varying obstacle fissure angles (α = 0–75°) and distances (d = 0.02–0.06 m). The results show that the obstacle fissure angles significantly regulate the crack paths: as the α increases, the tensile stress concentration shifts from the obstacle fissure’s middle to its ends, causing cracks to deflect toward the lower end, with a reduced propagation length and lapping time; at an α = 75°, the obstacle fissure’s lower tip dominates failure, forming an “induced fissure–lower end of obstacle fissure–top” penetration mode. The fissure distances affect the stress superposition: a smaller d (e.g., 0.02 m) induces vertical propagation and rapid lapping with the obstacle fissure’s lower end, while a larger d (e.g., 0.06 m) weakens the stress at the induced fissure tip, promoting horizontal deflection toward the obstacle fissure’s upper end and transforming the failure into “upper-end dominated.” This confirms that the improved SPH method effectively simulates crack behaviors, providing insights into multi-fissure failure mechanisms and theoretical support for hydraulic structure crack control and safety evaluation.

Temperature control is needed in the construction process of massive concrete and it can avoid the concrete cracks. Prediction of temperature development based on a hydration kinetics model can reduce the need for adiabatic temperature rise tests for concrete. However, the existing hydration kinetics model cannot accurately describe the hydration process of cement, thereby limiting the ability to further accurately predict the temperature rise of concrete based on the hydration kinetics model. This paper aims to establish a new hydration kinetics model, which is based on nucleation and growth model, and to predict the temperature development of concrete with set-controlling admixture based on this model. In this paper, the nucleation and growth of hydration products and the diffusion of free water by the modified boundary of nucleation and growth (BNG) model and the modified Fuji and Kondo’s model are described. The relationship between nucleation rate and apparent activation energy and the relationship between effective diffusion coefficient and apparent activation energy are linear. However, the relationship between growth rate and apparent activation is exponential. Finally, the temperature development of concrete can be calculated by the hydration degree of the cement.