Rigid polyurethane foam, foam concrete, and vacuum insulation board are common roofing insulation materials. Their weathering performance under long-term multi-field coupling determines the overall service life of the roof. The weathering properties of rigid polyurethane foam, foam concrete and vacuum insulation panels were studied under freeze thaw, humid-heat, dry-wet, high-low temperature, and multi-field coupling cycles, respectively. The heat transfer and construction process of roof panels was simulated base on upper loading and moisture transfer factors. The result indicates that the mass loss of the foam concrete and the rigid polyurethane foam in the weathering test was significant, which led to the gradual increase of thermal conductivity. Meanwhile, the thermal conductivity and mass loss of vacuum insulation panels did not change due to the lack of penetration under external pressure, therefore, it is necessary to construct composite thermal–insulation materials to alleviate the adverse effects of the service environment on a single material and realize the complementary advantages and disadvantages of the two materials. The results of the numerical simulations indicated that the roof structure must be waterproofed, and its weatherproof performance index should be the same as that of the thermal insulation material. Considering structural deformation, the overall heat transfer performance of the product was increased by around 5%.

Structural cracks in concrete have a significant influence on structural safety, so it is necessary to detect and monitor concrete cracks. Deep learning is a powerful tool for detecting cracks in concrete structures. However, it requires a large quantity of training samples and is costly in terms of computational time. In order to solve these difficulties, a deep learning target detection framework combining texture features with concrete crack data is proposed. Texture features and pre-processed concrete data are merged to increase the number of feature channels in order to reduce the demand of training samples for the model and improve training speed. With this framework, concrete crack detection can be realized even with a limited number of samples. To accomplish this aim, self-made steel fiber reinforced concrete crack data is used for comparison between our framework and those without texture feature mergence or pre-processed concrete data. The experimental results show that the number of parameters that need to be fitted in the model training and training time can be correspondingly reduced and the detection accuracy can also be improved.

Buildings consume many resources and generate greenhouse gases during construction. One of the main sources of greenhouse gases is carbon emission associated with buildings. This research is based on the computing rule of carbon emission at the materialization stage. By taking the features of green construction into consideration, quantitative analysis on construction carbon emission was undertaken via Life Cycle Assessment (LCA). Making use of Vensim (a system dynamics software package), we analyzed the amount of carbon emission at the materialization stage and determined the major subsystems affecting the carbon emission, then took into comprehensive consideration the differences of each subsystem’s carbon emission under different construction technologies. Under the mechanism of carbon trade at the materialization stage, the total price of carbon trades remains unchanged, while the trading price of each subsystem is adjusted. Under these conditions, a coefficient for step-wise increases in carbon price was proposed. By establishing such a system of gradient prices, construction companies are encouraged to adopt high-efficiency emission reduction technologies. Meanwhile, the system also provides a reference for the formulation of price-based policies about buildings’ carbon trading, and accelerates the process of energy conservation and emission reduction in China and the world at large.

Calcined coal gangue powder was used to partially replace Portland cement (PC) at 10%, 20%, 30%, and 40% by weight. The water to binder ratio (w/b) of 0.5 was used for all the blended cement paste mixes. The reaction degree of calcined coal gangue powder in calcined coal gangue powder/cement (AC) was determined using a selective dissolution method. It was found that high-volume calcined coal gangue powder pastes underwent a lower degree of reaction of calcined coal gangue powder. The effect of calcined coal gangue powder replacement on the reaction degree of calcined coal gangue powder in AC pastes in later age was less than that of in early age.

Chloride ion erosion is an important factor affecting the durability of marine engineering concrete. In particular, concrete structures in wave splash and tidal zones are subjected to dry and wet cycles and multidimensional diffusion of chloride ions. To investigate the intricate diffusion of chloride ions within concrete under these dynamic conditions, we devised a comprehensive experiment. This experiment encompasses multiple dimensions, involving dry and wet cycles, as well as static immersion. The experiment intends to reveal how chloride ions are distributed in the concrete and clarify the changes that occur in its microstructure. Based on Fick’s second law, the multidimensional diffusion model of chloride ions in concrete under the dry and wet cycles and static immersion was established by comprehensively considering the effects of chloride ion exposure time, environment temperature, relative humidity, and the action of dry and wet cycles. The results show that, under the same conditions, the chloride content in concrete decreases with the increase in penetration depth but increases with the increase in the chloride diffusion dimension and exposure time. Dry and wet cycles and multidimensional diffusion of chloride ions increase the development of cracks and pores in the concrete structure and generate large quantities of C₃A·CaCl₂·10H₂O, which will exacerbate the chloride ion transport rate and penetration depth of concrete. Under the same exposure time and penetration depth, the chloride ion content in concrete under two-dimensional (2D) and three-dimensional (3D) diffusion under dry and wet cycles was 1.09~4.08 times higher than that under one-dimensional (1D) diffusion. The correlation coefficients between the simulation results of the multidimensional transport model of chloride ions in concrete under multi-factor coupling and the experimental results were all greater than 0.95, and the model can be utilized to predict the distribution of chloride ion concentration in concrete.