In order to better study the actual stress state of the tunnel pipe-roof, according to the tunnel construction process and construction influence, the lining segment of the pipe roof tunnel was divided into five segments: the second lining section, the primary support section, the excavation section, and the disturbed and undisturbed section in front of the palm face. Considering the whole pipe-roof arch effect formed after pipe-roof grouting, a Winkler elastic foundation beam model of the five-sections tunnel pipe-roof is established. The flexure and internal force calculation formula of the five-paragraph tunnel pipe-roof model are deduced and compared with existing engineering cases. At the same time, different excavation footage, pipe diameters and elastic modulus of reinforced area are analyzed. The results show that: The maximum deflection of the pipe-roof obtained by the new model and the corresponding position are closer to the on-site monitoring results, which can solve the problem of pipe-roof deformation in the disturbance section caused by the unreasonable setting of boundary conditions in the existing pipe-roof model. The arching phenomenon at the end is more in line with the actual deformation of the pipe-roof. Changing the excavation footage has a greater impact on the deflection of the pipe-roof. When the diameter of steel pipe is 108 mm and 114 mm, the pipe-roof support effect are similar; when E_c(elastic modulus of converging) is greater than 40.0 E_g(elastic modulus of ground rock), the excavation footage and the diameter of the steel pipe have little effect on the deflection and bending moment of the pipe-roof. Therefore, increasing the elastic modulus of the reinforced area is the most effective method for reducing pipe-roof deflection and internal force.

In order to judge the stability of surrounding rock in loess tunnel construction more correctly and reasonably, a statistical constitutive model of loess damage is proposed, and a limit strain method is proposed to judge the stability of tunnel construction. In this paper, it is assumed that the strength of loess micro elements is subject to composite power function, and the continuum damage mechanics theory and equivalent strain theory are used as the modeling basis. Meanwhile, considering the influence of confining pressure and water content on the deformation modulus of loess, a new constitutive model which can reflect the stress-strain relationship of loess under different water content and different confining pressure is established by using statistical method. The results show that the model established in this paper can well reflect the strain hardening characteristics of loess during the whole loading process, especially the strain softening characteristics under low confining pressure. Based on the statistical constitutive model of loess damage established in this paper, the formula for determining the limit strain of surrounding rock is modified on the basis of the original method for determining the limit strain of surrounding rock. Finally, a method for determining the stability of tunnel construction in loess stratum is proposed and applied to practical engineering.

The rapid development of social economy leads to an increasingly obvious contradiction between transport demand and capacity. The reconstruction and expansion of existing highway tunnels have become a long-term and effective means to improve the highway transportation environment in the future. Through the literature retrieval and case statistical analysis of typical reconstruction and expansion of highway tunnels, the development trend and research trends of highway tunnel reconstruction and expansion are clarified. Point out the research status and deficiencies of highway tunnel reconstruction and expansion schemes, and propose reconstruction and expansion schemes based on the comparison and selection of schemes. Combined with four typical reconstruction and expansion construction methods and tunnel expansion coefficients, and construction methods for reconstruction and expansion of existing highway tunnels are put forward. The research dynamics and deficiencies of surrounding rock pressure calculation, surrounding rock mechanical response, blasting vibration control and supporting parameters of the reconstruction and expansion highway tunnel are expounded. The application prospects of BIM technology, artificial intelligence technology, and green carbon neutral concept in the reconstruction and expansion of tunnels are put forward, and it is pointed out that the standardization of construction standards, the intelligentization of expansion plans, the innovation of construction methods and the systematization of theoretical research are the key research directions in the future.

Due to the huge investment of tunnel boring machine (TBM) in the early stage, it is of great significance to reasonably classify the excavability of TBM surrounding rock and give the suggested values of construction parameters for predicting the construction period and controlling the cost. The TBM construction project of Yangtaishan hard granite tunnel in the section 6101 of Shenzhen Metro Line 6 between Dalang station and Shiyan Station is taken as the research background. Firstly, the tunneling penetration index (FPI) is taken as the tunneling performance evaluation index, and Spearman correlation was used to analyze the rationality of FPI as an evaluation index. On this basis, FPI is used as the input parameter to establish the penetration degree prediction model, cutter head thrust prediction model and driving speed prediction model, and the reliability of the models is verified based on the right line driving data of Yangtaishan tunnel. Based on the K-means clustering method and FPI clustering index, the excavation grade of surrounding rock was divided. According to the prediction formula of tunneling parameters given in this paper, the recommended values of tunneling parameters corresponding to the excavation grade were given. The results show that: (1) FPI can reasonably reflect rock-mechanism system in TBM construction to a certain extent, and can be used as an evaluation index of tunneling performance. (2) The prediction formula of TBM tunneling parameters established in this paper has a good correlation coefficient R² of 0.75, 0.98 and 0.97, which can accurately provide prediction parameters for TBM tunneling. (3) The classification model of TBM surrounding rock excavability established in this paper divides surrounding rock into five grades, and the tunneling parameter values under different levels are determined.