The stratum in coastal area is weak and rich in water with low bearing capacity. During the engineering construction by frozen method, the deformation and temperature fields of the structure and soil respond violently, which can easily lead to large-scale construction disturbances. Temperature effect parameters of soil and groundwater are obtained through field and indoor tests, a 3D thermo-mechanical coupled numerical analysis model for the whole process of freezing excavation thawing is established, the development and diffusion laws of the frozen curtain ring during the construction process is explored, the deformation and temperature evolution characteristics of soil and support structures are revealed, and the research results are finally applied and verified through engineering practice. The results show that: During the channel excavation, the soil within a range of 2.75~7.16 m from the channel center is compressed and compacted, and there are sharp temperature fluctuations. Within 15~25 days of heating and thawing, grouting reinforcement can be carried out on the soil within a depth range of 15.8 m below the ground surface. Especially within a range of 3.99 m from the channel center, the soil temperature rise faster than other regions. During the soil freezing process, the soil on the inner side of the freezing tube freezes faster than the outer side, and the soil on both sides of the channel freezes faster than the upper and lower sides. During the heating and thawing process, the upper and lower soil in the channel thaw faster than the two sides, and the thawing of the upper and lower soil in the channel mainly occurs during the continuous heating stage. During the thawing process, the deformation of the upper and lower parts of the primary support and secondary lining is significantly greater than that of the two sides, and there will be significant stress concentration at the corners of the steel rib bottom and the top and lower parts of the reinforcement mesh. Pre-embedded grouting pipes can be used to inject grout into the upper and lower soil 20 to 40 days after thawing begins, in order to slow down the development of structural displacement and stress concentration.

With the rapid construction of highway tunnels in China turns to the stage of both construction and maintenance, it will face the double pressure of the rapid growth of operating mileage and the deterioration of existing tunnels in the future, and the mobile detection and rapid diagnosis of structural safety have become research hotspots in the field of highway tunnel operation and maintenance. China has developed a lot of detection vehicles being used to tunnel apparent defects, which can provide an efficient method for rapid detection of tunnel surface cracks, water leakage, etc. However, highway tunnel lining images are characterized by complex backgrounds, multiple interference factors and low cracks occupancy, and accordingly, it’s difficult to do rapid identification and analysis of mass tunnel detection data, which has become the main bottleneck of technology promotion. Based on deep learning algorithm, this paper proposes a method that combines the target recognition with semantic segmentation algorithm. Firstly, Faster R-CNN deep neural network is used for target recognition on the original lining images to determine whether there are cracks and intelligently mark the crack regions within a rectangle. Then, the crack area selected by the frame is automatically cut, thereby filtering the pictures without cracks and removing the interference background in the pictures containing cracks, and then using the U-Net semantic segmentation network to segment the cracks at the pixel level. In practical engineering, it takes less than 0.15s to recognize the cracks in one image. Besides, this method can effectively identify various cracks in complex environments, with the F1 score of target recognition reaching 92% and the segmentation accuracy rate 98%. Compared with the previous feature recognition and global segmentation intelligent algorithms, this method significantly improves the speed and accuracy of recognition, and provides a feasible method for rapid and accurate recognition of mass defect detection data.

In the context of the increasing number of tunnel diseases, ensuring safe and efficient operation throughout the tunnel life is an issue that must be faced now and in the future. Unlike other buildings or equipment, tunnels are a linear longitudinal, self-similar transversely, and buried underground, making it difficult to apply existing maintenance strategies to tunnel structures. For prefabricated assembled tunnels, engineering research and theoretical analysis were used to propose a suitable forward-looking maintenance strategy for the tunnels. The research includes: firstly, summarizing the types and causes of prefabricated assembled tunnel diseases, proposing a five-layer disease classification method and a three-world, eight-door disease cause classification method. Then, the special characteristics of tunnel structure and diseases are analyzed to reveal the special characteristics of tunnel maintenance. Then, the prerequisites of maintenance and the purpose of maintenance are analyzed from the tunnel service requirements and perspectives. Then, the macro dynamic evolution mechanism of prefabricated assembled tunnel disease is proposed based on the classification of the three boundaries and the eight causes of disease. Finally, a pre-collective maintenance strategy applicable to tunnels is proposed based on the evolutionary mechanism and maintenance specificity. The research results are not only applicable to prefabricated assembled tunnels, but also to cast-in-place tunnels, and also have theoretical guidance and engineering reference significance for underground structure-type projects.

The overload on the ground surface causes the longitudinally uneven settlement of tunnels and endangers the safety of tunnel operations. The settlement monitoring data of tunnels plays a significant role in analyzing the safety of the tunnels. A method of tunnel settlement prediction based on Persistent Scatterer Interferometric Synthetic Aperture Radar (PS-InSAR) under ground surface overload is proposed. The method is applied in the Shanghai Yanggaozhong Road power tunnel project. The results show that: PS-InSAR technology can be applied to collect high-accuracy ground settlement data with an average error of 0.78mm/y compared with ground-level monitoring data. The tunnel settlement rate curve predicted by this method is relatively consistent with the tunnel level measured curve, and the location of the longitudinally uneven settlement of the tunnel is detected accurately. Compared to the measured settlement rate of the tunnel, the average error of predicted settlement rate for the tunnel is 3.09mm/y. The main factors affecting the accuracy of the results include the fitting error of ground surface settlement caused by using PS-InSAR data, the settlement calculation error of soil layers above the tunnel, and the soil disturbance caused by the construction activities in the vicinity of the tunnel. PS-InSAR shows such advantages as the wide range of monitoring and the access to historical data, which contributes new ideas to the acquisition and analysis of tunnel settlement data in soft soil areas.