Bochao Sun

Under high‐load conditions, accurate strain prediction remains challenging due to nonlinear responses, noise, and incomplete measurements. This study proposes a pointwise dual‐branch neural regression model using multisource data. A 1:8 scale‐reinforced concrete frame was modeled in ABAQUS, and based on measurements up to 1.35 times the design load, three loading conditions were simulated to generate finite element (FE) strain data. The model combines an encoder‐style nonlinear branch implemented using a Transformer encoder block with an MLP branch representing the dominant load–strain trend, while a dynamically weighted loss increases emphasis on large‐strain samples. The best tested configurations achieve validation R ² values of 0.9975 and 0.9997, with high‐load MAPE values of 2.56% for experimental S7 extrapolation and 0.39% for held F5‐at‐1.0× cross‐condition FE prediction. Under high‐noise conditions, multisource data fusion increases validation R ² from 0.9621 to 0.9965 while reducing high‐load MAPE from 11.15% to 3.09%. These results demonstrate stable prediction performance within the tested benchmark.

The application of explainable artificial intelligence (XAI) in civil engineering has garnered increasing attention due to its ability to enhance transparency in machine learning (ML) models for mater...

The global outbreak and prolonged presence of Coronavirus Disease 2019 (COVID-19) have resulted in a substantial accumulation of discarded masks, posing serious environmental challenges. This study proposes an eco-friendly and low-carbon strategy to repurpose discarded DMFM fibers as a key component in fiber-reinforced self-compacting recycled aggregate concrete (FRSCRAC). The mechanical and environmental performance of FRSCRAC was systematically evaluated by investigating the effects of recycled coarse aggregate (RCA) replacement ratios (0%, 50%, 100%), discarded DMFM fiber material (DMFM) contents (0%, 0.1%, 0.2%, 0.3%), and fiber lengths (2 cm, 3 cm, 4 cm) on axial compression failure mode and stress–strain behavior. The results demonstrated that DMFM fibers significantly enhanced concrete ductility and peak stress via the fiber-bridging effect. Based on fiber influence, modified stress–strain and shrinkage models for SCRAC were established. To further understand the fiber fixation mechanism, X-ray computed tomography (X-CT) and scanning electron microscopy (SEM) analyses were conducted. The findings revealed a stable random distribution of fibers and strong interfacial bonding between fibers. These improvements contributed to enhanced mechanical performance and the effective immobilization of polypropylene microfibers, preventing further microplastics release into the air. This innovative approach provides a sustainable solution for recycling and effectively immobilizing discarded DMFM fibers in concrete over long curing periods, while also enhancing its properties.

Fires frequently affect building structures, yet the mechanical behavior of deep flexural members, especially after exposure to high temperatures, remains inadequately understood. Current design codes for concrete structures at ambient temperatures do not sufficiently account for the residual capacity of these members following fire‐induced damage. Given the specific and complex mechanical properties of deep flexural members, particularly under varying shear span and shear‐depth ratios, a gap exists in predicting their post‐fire performance. Addressing this gap is critical to improving the safety and resilience of fire‐exposed structures. Here, the shear behavior of eight deep flexural members was experimentally investigated at both normal temperatures and post‐high‐temperature conditions (200°C, 350°C, and 500°C). The specimens were subjected to tests with shear span ratios between 0.3 and 1.2, and shear‐to‐depth ratios of 2 and 3. Key performance metrics—including failure modes, ultimate bearing capacity, load‐deflection relationships, steel strain, and maximum crack width—were analyzed. Additionally, the empirical formula in existing design codes was modified using a layered method to better predict the residual shear capacity of deep flexural members after fire exposure. The results demonstrate that high temperatures and shear span ratios significantly affect shear performance. Shear capacity initially improved and then declined as temperature increased. A higher shear span ratio intensified bending effects, transitioning failure modes from shear failure to bending‐shear failure, and progressively reduced ultimate bearing capacity. The proposed empirical formula provides a reliable means of calculating or predicting the residual bearing capacity of fire‐damaged deep flexural members, offering sufficient safety margins. These findings contribute to refining post‐fire structural assessments and enhancing building safety.