With the continuous development of production technology, the performance of glass-fiber-reinforced polymer (GFRP) bars is also changing, and some design codes are no longer applicable to new materials based on previous research results. In this study, a series of durability tests were carried out on a new generation of GFRP bars in laboratory-simulated seawater and a concrete environment under different temperatures and sustained loads. The durability performance of GFRP bars was investigated by analysing the residual tensile properties. The degradation mechanism of GFRP bars was also analysed by scanning electronic microscopy (SEM). Furthermore, the long-term performance of GFRP bars exposed to concrete pore solution under different stress levels was predicted using Arrhenius theory. The research results show that the degradation rate of GFRP bars was increased significantly at a 40% stress level. By comparing the test results, design limits, and other scholars’ research results, it is demonstrated that the GFRP bars used in this test have a good durability performance. It is found that the main degradation mechanism of the GFRP bars is the debonding at the fiber-matrix interface. In the range test, the effects of a 20% stress level on the degradation of GFRP bars were not obvious. However, the long-term performance prediction results show that when the exposure time was long enough, the degradation processes were accelerated by a 20% stress level.

Structural cracks significantly threaten the safety and longevity of historical buildings, which are essential parts of cultural heritage. Conventional inspection techniques, which depend heavily on manual visual evaluations, tend to be inefficient and subjective. This research introduces an automated framework for crack and damage detection using advanced YOLO (You Only Look Once) models, aiming to improve both the accuracy and efficiency of monitoring heritage structures. A dataset comprising 2500 high-resolution images was gathered from historical buildings and categorized into four levels of damage: no damage, minor, moderate, and severe. Following preprocessing and data augmentation, a total of 5000 labeled images were utilized to train and evaluate four YOLO variants: YOLOv5, YOLOv8, YOLOv10, and YOLOv11. The models’ performances were measured using metrics such as precision, recall, mAP@50, mAP@50–95, as well as losses related to bounding box regression, classification, and distribution. Experimental findings reveal that YOLOv10 surpasses other models in multi-target detection and identifying minor damage, achieving higher localization accuracy and faster inference speeds. YOLOv8 and YOLOv11 demonstrate consistent performance and strong adaptability, whereas YOLOv5 converges rapidly but shows weaker validation results. Further testing confirms YOLOv10’s effectiveness across different structural components, including walls, beams, and ceilings. This study highlights the practicality of deep learning-based crack detection methods for preserving building heritage. Future advancements could include combining semantic segmentation networks (e.g., U-Net) with attention mechanisms to further refine detection accuracy in complex scenarios.

Since glass fiber-reinforced polymer (GFRP) bars have a lower modulus than steel bars, the design of GFRP-reinforced concrete (GFRP-RC) is often governed by the serviceability limit state (deflection and cracking) rather than the ultimate state. A new design method has been proposed in this paper for GFRP-RC beams based on the flexure crack width. The state when the maximum flexure crack width in the tensile zone reaches the limit of 0.5 mm specified by ACI 440.1R-15 was used as the design limit state. The concrete compressive strain at the extreme compression fiber of concrete under the design limit state was obtained by four-point bending tests of eight full-scale GFRP-RC beams and finite element analysis. Based on the concrete compressive strain under the design limit state and cross-sectional analysis, a design method for calculating the longitudinal reinforcement ratio of GFRP-RC beams under the design limit state is proposed. This design method is proven to be feasible by the experimental and the finite element results. In addition, the flexural capacity coefficient was discussed to investigate the safety reserve of the design method.

Data-driven fault diagnosis for rotating machinery faces challenges in deep feature extraction, noise interference, and hyperparameter sensitivity. To address these issues, we propose a novel deep learning framework integrating three key innovations: (1) A multi-channel parallel network combining CNN and TCN to extract spatiotemporal features via dense connections, preventing gradient vanishing; (2) An adaptive soft-threshold denoising module driven by self-attention mechanisms, dynamically suppressing noise while preserving critical fault signatures; (3) A Bayesian hyperparameter optimizer automating model configuration, significantly reducing manual tuning efforts. Additionally, a penalty-enhanced loss function is designed to improve classification of hard-to-distinguish faults. Extensive experiments on two bearing datasets (CWRU and WHUT) demonstrate superior performance: 98.84% accuracy on CWRU and 99.89% on WHUT, outperforming ResNet, DenseNet, and other benchmarks by 5%–30%. Especially, it outperforms the benchmark by 3%–50% in noise conditions. The framework shows strong potential for industrial predictive maintenance under noisy environments.

Reconstructing 3D building models from point clouds acquired by UAV sensors remains challenging due to irregular building geometries and sensor noise. This paper proposes an unsupervised, geometry-oriented reconstruction method based on the line-symmetric bilateral point distribution features. The method constructs baselines from farthest point pairs within local bounding spheres, then applies dual-parameter constraints (point–line distance and point statistics on both sides of a line of symmetry) combined with density peak ranking to detect building corners without training data. Evaluations show that the method reduces Cloud-to-Mesh error by 15–20% over Polyfit, DIF method, and PolyGNN, while retaining fine details in complex L-shaped and U-shaped buildings. Reconstruction quality remains stable under 0.05 m Gaussian noise. The training-free, lightweight design enables scalable, automated 3D building reconstruction for digital-twin applications.

In densely built urban environments, GNSS signals frequently undergo diffraction at building edges, and the resulting errors can severely degrade positioning accuracy and reliability. Previous studies have shown a strong correlation between diffraction error and the carrier-power-to-noise-density ratio (C/N0). Building on this observation, this study proposes a GNSS diffraction-mitigation method based on segmented down-weighting and exclusion of affected observations. First, an open-sky reference model of the elevation–C/N0 relationship is established for each satellite class. A robust strategy is then introduced to adaptively down-weight moderately contaminated observations and remove severely affected ones during stochastic modeling. The proposed method is evaluated using both static and kinematic datasets collected in dense urban environments. In the static experiment under severe building obstruction, the ambiguity-fixing rate (AFR) reaches 95.5%, with horizontal and vertical accuracies of 4 mm and 8 mm, respectively, substantially outperforming conventional weighting strategies. In the vehicle-based kinematic experiment, the fixed-solution rate exceeds 80%, and the float solution is also noticeably improved relative to traditional weighting and exclusion methods. Overall, the proposed method effectively mitigates diffraction-induced errors and improves positioning performance in dense urban environments, with potential applications in automated inspection, intelligent construction, and high-precision deformation monitoring.