Within the framework of intelligent bridge detection, a number of crack detection methods based on image processing techniques have been implemented. In this study, a combined novel approach with deep learning of a single shot multibox detector (SSD) and the eight neighborhood algorithm is proposed and applied to bridge crack image identification to provide an automatic method for crack detection. First, a large number of concrete crack images collected from the site were segmented and preprocessed for the establishment of a crack image dataset. Deep learning of the SSD algorithm was introduced on the training set to establish the detection model, where the model parameters were adjusted by the validation set. Sliding window technology was integrated to identify the cracks in the test set. The effects of the sliding window size and dataset size on the crack detection results were discussed. Moreover, the eight neighborhood algorithm was adopted for further crack detection correction. The results show that the configuration achieves good crack detection by the deep learning of the SSD algorithm with high precision and recall. The introduction of the eight neighborhood correction algorithm further improves the detection results by eliminating some misjudged results. Finally, the developed algorithm was placed into a portable device, with which cracks were effectively identified. The introduced method shows significantly better performance in crack detection, and the system installed on the portable device provides a way to broaden its application in the automatic crack detection of concrete bridges.

Non-destructive analysis of fiber reinforced polymer (FRP) composites is important for confirming the long-term safety and durability of concrete structures. In this study, a pulse-heating infrared thermography technique was used to detect and characterize bonding defects of externally bonded carbon fiber reinforced polymers (CFRP) on concrete surface structures. The CFRP composite contains various bonding defects of three different sizes located at five different depths. Sequential thermal images were obtained to describe the temperature contrast and shapes of the bonding defects. Through analysis of the maximum temperature response, we investigated the effects of defect size and depth on the defect temperature response. The relationship between the defect depth and maximum temperature response was used to quantitatively estimate the defect depth. In addition, finite element simulations were performed on the CFRP composites with bonding defects to investigate the temperature response of various defects, which showed good agreement with the experimental results. This confirms the effectiveness of the infrared thermography method to detect and characterize bonding defects of FRP composites bonded on concrete structures.

Permeability, as a critical element in making sure the underground facilities are secure, is a vital consideration in analyzing the rock material seepage. Selecting granite gneiss from underground oil storage as the research sample in this study. Triaxial mechanical property tests under different hydro-mechanical properties were carried out under the rock full-automatic triaxial servo system that controls the application of axial pressure, confining pressure, and seepage pressure. Through experiments carried out, we obtained the rock samples’ mechanical properties and permeability in three stages of the stress–strain process. The study shows that the seepage pressure considerably affects granite gneiss strength and deformation parameters under hydro-mechanical properties. On the basis of the same confining pressure, in pace with the growth in seepage pressure, elastic modulus, the deformation modulus, and the peak strength present a prominent decreasing inclination. The derived mechanical parameters are bound up with the stages we divided. This study analyzes and discusses the relationship among the strains and permeability, establishing the granite gneiss hydro-mechanical coupling constitutive model. Verification shows the results in numerical and experimental matches well, indicating that the rock hydro-mechanical properties could be effectively represented by the constitutive model.

The addition of a conductive material into polymer improves its mechanical properties, electrical properties and thermal conductivity and bestows it with good self-sensing and self-adjusting properties. In this study, carbon nanotubes-modified epoxy resins (CNTs-EP) were successfully prepared with good dispersion through the combined methods of three roller rolling, ultrasonic processing and adding surfactant. Tests were conducted to evaluate the resistivity of unloaded modified epoxy resins with different mixing amounts of carbon nanotubes (CNTs), to determine the conductive percolation threshold. On the basis of the test results, a series of monotonic and cyclic uniaxial tensile tests were then conducted to investigate the force sensitivity of the conductivity of epoxy resins modified with different mixing amounts of CNTs. The relationship between the stress and the resistivity under various mixing amounts was studied, indicating that the resistance response could play a good warning role on the damage of the modified polymer material.

Fatigue behavior is an important factor for mechanical analysis of concrete members reinforced by basalt fiber reinforced polymer (BFRP) grid and polymer cement mortar (PCM) and plays a critical role in ensuring the safety of reinforced concrete bridges and other structures. In this study, on the basis of the static loading test results of concrete specimens reinforced by BFRP grid and PCM, a series of fatigue tests with different loading levels were conducted on interfaces between BFRP grid and concrete to investigate the fatigue behavior of BFRP grid‐concrete interfaces. The test results indicate that with high loading level, the fatigue failure mode of interface is interfacial peeling failure while it transforms to the fatigue fracture of the BFRP grid under low loading level. The fatigue life (S-N) curves of BFRP grid‐concrete interface are obtained and fitted in stages according to different failure modes, and the critical point of the two failure modes is pointed out. The relative slip evolution of interface during fatigue is further revealed in different stages with two failure modes, and the law of interface strain is studied with the increase of fatigue times. The relation of effective bonding length of interface and fatigue times is also described.