Search

Abstract As construction technology and project management develop, structural monitoring systems become increasingly important for ensuring large-span spatial structure safety during construction and operation. However, most of the sensors and monitoring equipment in monitoring systems are poorly serviced, resulting in frequent abnormal monitoring data, which directly leads to challenges in data analysis and structural safety assessment. In this paper, a structural response recovery method based on a long short-term memory (LSTM) neural network is proposed by studying the autocorrelation of data and the spatial correlations among data at multiple measurement points. The effectiveness and robustness of the proposed method are verified using the monitored stress data for a grid structure jacking construction process, and the influence of different data loss rates on the recovery accuracy is analysed. The recovery models are compared using a support vector machine and a Multi-Layer Perception (MLP) neural network. The proposed method can effectively restore missing data; notably, the MSE index is 0.6, and the MAPE is below 15%. The data restoration method based on the LSTM neural network is more accurate than the traditional method. Finally, the repair applicability of various types of monitored data is verified using the monitoring data from Hall F of Qingdao Jiao-dong International Airport under typhoon conditions.

This paper introduces an enhanced you only look once (YOLO) v5s‐D network customized for detecting various categories of damage to post‐fire reinforced concrete (RC) components. These damage types encompass surface soot, cracks, concrete spalling, and rebar exposure. A dataset containing 1536 images depicting damaged RC components was compiled. By integrating ShuffleNet, adaptive attention mechanisms, and a feature enhancement module, the capability of the network for multi‐scale feature extraction in complex backgrounds was improved, alongside a reduction in model parameters. Consequently, YOLOv5s‐D achieved a detection accuracy of 93%, marking an 11% enhancement over the baseline YOLOv5s network. Comparison and ablation tests conducted on different modules, varying dataset sizes, against other state‐of‐the‐art networks, and on public datasets validate the resilience, superiority, and generalization capability of YOLOv5s‐D. Finally, an application leveraging YOLOv5s‐D was developed and integrated into a mobile device to facilitate real‐time detection of post‐fire damaged RC components. This application can integrate diverse fire scenarios and data types, expanding its scope in future. The proposed detection method compensates for the subjective limitations of manual inspections, providing a reference for damage assessment. [ABSTRACT FROM AUTHOR]

To obtain a precise finite-element model (FEM) for analyzing the dynamic response of corroded beams at high temperatures, a stepwise FEM modification strategy is proposed based on the improved extreme learning machine. Three concrete beams were designed and cast, and the dynamic response characteristics of corroded concrete beams at room temperature and high temperature are discussed. Firstly, electrical accelerated corrosion tests and vibration tests were conducted on simply supported beams at room temperature. The fundamental frequencies of concrete beams under different corrosion ratios were measured. The attenuation law of fundamental frequency with corrosion ratio also was studied. Subsequently, the FEM under different corrosion ratios was modified. The bond-slip between steel bars and concrete under different degrees of corrosion was considered during the correction process. Finally, a vibration test at high temperature was performed. The modal attenuation law of corroded beams at high temperatures was analyzed. Based on the modified FEM, numerical analysis at high temperature was performed. The proposed FEM modification strategy and the study of the attenuation regularities of modal information under fire exposure provide a foundation for further research on the damage development of corroded reinforced concrete (RC) beams under fire exposure. [ABSTRACT FROM AUTHOR]

In order to study the effect of high temperatures on the thermomechanical properties of reinforced concrete (RC) beams, fire and mechanical tests were carried out on four fully dimensioned simply supported beams. A numerical analysis model based on the measured information was developed, and validated through experimental tests and theoretical calculations. The effects of fire exposure time, different physical parameters and concrete spalling parameters on the thermodynamic properties were further analysed. The results show that the measured vibration frequencies exhibit a fluctuating trend of decay with increasing fire exposure time. The simulated frequency decays as a power function of fire exposure time. The frequency decreases with increasing span-to-height ratio and increases with section width, elasticity modulus of the concrete and reinforcement ratio. Cross-sectional temperature field is most significantly affected by spalling depth, and increases rapidly with an increase in this parameter. With the increase in spalling area ratio, the increase in the temperature field gradually becomes smaller. The effect of spalling area ratio was relatively stable after more than 15%. Flexural bearing capacity shows a linear trend of decrease with increase in spalling depth. With the increase in spalling area ratio, it shows a power function of the decreasing trend. [ABSTRACT FROM AUTHOR]

The fire resistance of corroded reinforced concrete has attracted extensive attention, mainly due to the frequent occurrence of fires in old structures. However, using the postfire bond strength to predict structural capacity during fires leads to inaccurate results. Thus, in this study, an accelerated corrosion test and an eccentric pullout test were performed under steady-state conditions from 20°C to 800°C. The test results indicated that compared with the bond strength at 20°C, the bond strength at 200°C decreased by 6% for the corroded specimen (corrosion level below 4.3%) and increased by 11% for the uncorroded specimen. When the temperature exceeded 500°C, the storage modulus and bond strength decreased 55%. Increasing the rebar diameter reduced the bond strength by approximately 29% and made the descending section of the τ-s curve steeper during exposure to fire. The energy dissipation at 100°C was larger than that at other temperatures, and the bond stiffness at 600°C was 12.5% of that at 20°C. The reduction in bond stiffness postponed the development of splitting cracks, restricted the crack width, and prevented splitting failure. Subsequently, temperatures dependence was analyzed and methods were proposed for the calculation of bond strengths at elevated temperatures. The results obtained with the proposed continuous bond-slip model were in satisfactory agreement with the pullout test results. Finally, the comparison of integral absolute error indicated that the model could be used to make accurate calculations of bond strengths during fires. [ABSTRACT FROM AUTHOR]

The bond performance of corroded reinforced concrete (RC) subjected to fire has a significant influence on evaluating the structural load-bearing capacity and early warning for structure collapse. However, the bond deterioration mechanism of corroded RC subjected to fire has not been sufficiently studied. Therefore, in this study, the bond behavior between corroded reinforcement and concrete at elevated temperatures was systematically investigated by testing 76 corroded specimens obtained by the accelerated corrosion method. Eccentric pull-out tests were performed immediately when the temperature at the interface of the reinforcement and concrete reached the target values. The main parameters include the degree of corrosion (i.e., 0%, 2%, 5%, 10%, and 20%), concrete cover thickness (20, 30, and 40 mm), and temperature (20°C, 100°C, 200°C, 400°C, 500°C, 600°C, and 800°C). The test results indicated that when exposed to temperatures below 400°C, the bond strength at elevated temperatures is less than that after natural cooling. The bond strength changed minimally at elevated temperatures between 200°C and 400°C, while a significant reduction was observed after natural cooling. Moreover, the bond strength increased by 8.7% at less than 3.4% corrosion degree. However, severe corrosion (with a corrosion degree exceeding 13.2%) and high temperature (exceeding 500°C) decreased the bond strength by 16.3% and 29.6%, respectively. In addition, methods for calculating the bond strength and residual bond strength based on the deterioration of concrete compressive strength at elevated temperatures are proposed. [ABSTRACT FROM AUTHOR]

The use of medical image synthesis with generative adversarial networks (GAN) is effective for expanding medical samples. The structural consistency between the synthesized and actual image is a key indicator of the quality of the synthesized image, and the region of interest (ROI) of the synthesized image is related to its usability, and these parameters are the two key issues in image synthesis. In this paper, the fusion-ROI patch GAN (Fproi-GAN) model was constructed by incorporating a priori regional feature based on the two-stage cycle consistency mechanism of cycleGAN. This model has improved the tissue contrast of ROI and achieved the pairwise synthesis of high-quality medical images and their corresponding ROIs. The quantitative evaluation results in two publicly available datasets, INbreast and BRATS 2017, show that the synthesized ROI images have a DICE coefficient of 0.981 ± 0.11 and a Hausdorff distance of 4.21 ± 2.84 relative to the original images. The classification experimental results show that the synthesized images can effectively assist in the training of machine learning models, improve the generalization performance of prediction models, and improve the classification accuracy by 4% and sensitivity by 5.3% compared with the cycleGAN method. Hence, the paired medical images synthesized using Fproi-GAN have high quality and structural consistency with real medical images. [ABSTRACT FROM AUTHOR]