The digital economy is an important engine in leading the high-quality development of the economy, and entrepreneurship is an endogenous force propelling economic growth. These factors are essential drivers for advancing the transformation and upgrading of the construction industry and improving efficiency. This study integrates entrepreneurship into the analytical framework of industrial development and explores the primary pathways and theoretical mechanisms of the digital economy for the high-quality development of the construction industry from the dimensions of entrepreneurial innovation spirit and entrepreneurial pioneering spirit. On this basis, we empirically test the role mechanism between the digital economy, entrepreneurship, and the high-quality development of the construction industry. The results of this study indicate that (1) the digital economy significantly promotes the high-quality development of the construction industry through an improvement in technical efficiency; (2) the innovation and entrepreneurship of entrepreneurs are important mechanisms for the digital economy to promote the high-quality development of the construction industry, and at this stage, entrepreneurial spirit promotes more than the spirit of innovation; and (3) the development of the digital economy plays a greater role in enhancing low-productivity regions and the central and eastern regions, and the inclusive characteristics of the digital economy are beginning to emerge.

Shield tunnels assembled with general ring segments are widely used in urban areas. Segment assembly methods and widths cause changes in the mechanical properties of the structure and influence the seismic response of shield tunnels. To investigate the influence of the assembly method and width of the general ring segment on the seismic performance of a shield tunnel, a three-dimensional refined soil–structure dynamic interaction finite element model of the shield tunnel was established based on ABAQUS, and the mechanical response and joint deformation of the general ring lining under seismic loads were studied. The simulation results show the following: (i) The overall deformation of the tunnel lining is not significantly affected by the assembly method, and the difference is only 5.24% under a 0.4 g earthquake. (ii) The seismic responses of general ring tunnels with different assembly methods are quite different, and the mechanical properties of the shield tunnel assembled with the straight assembly method are better than those of the shield tunnel assembled with staggered joints, but the deformation of the structure is larger. Under the action of a 0.1 g earthquake, the radial force, circumferential force, and bending moment of the staggered 90° assembly tunnel are respectively reduced by 13.6%, 11.1%, and 17.8% compared with the staggered 45° assembly structure, but the maximum intra-opening deformation increases by 0.19, 0.58, and 2.4 mm, respectively. (iii) The internal force distribution of the bolts is controlled by the deformation of the joint; compared with the CF90 and TF assembled tunnels, the mechanical properties and deformation characteristics of the CF45 and CF90 assembled tunnels are more reasonable. (iv) The extrados and intrados joint opening deformation and shear dislocation of the 1.2 m wide general ring segment under the staggered assembly increase by 1.2 mm and 1.03 mm, respectively, compared with the 1.5 m wide segment, while the radial force, circumferential force, and bending moment are reduced by 24.4%, 36.5%, and 41.7%, respectively, indicating that the seismic performance of the shield tunnel with a segment width of 1.5 m is better than that of the shield tunnel with a width of 1.2 m.

Due to the extremely strong rheological properties of soft clay, the dynamic characteristics of undisturbed saturated soft clay with sand layers in Ningbo area were studied. Experimental results demonstrate that the dynamic pore pressure, cumulative axial strain and double amplitude dynamic strain of unidirectional vibration cycle increase with the axial cyclic stress ratio increasing, and because the excess hydrostatic pore water pressure will gradually dissipate with vibration cycle increasing, the strain increases and the dynamic pore pressure decreases with the progress of vibration cycle. Besides, it can be noticed that the dynamic pore pressure and double amplitude dynamic strain of bidirectional vibration cycle are generally larger than that of unidirectional vibration cycle. Because bidirectional vibration cycle needs controlling the change of axial stress in the horizontal and vertical direction, larger instantaneous strain will be generated by external load, and excess hydrostatic pore water pressure is larger. In addition, due to the negative correlation between dynamic elastic modulus and strain, the dynamic elastic modulus of bidirectional vibration cycle is generally smaller than that of unidirectional vibration cycle. Finally, with the cyclic vibration continuing, energy is continuously lost and damping ratio decreases.

Current methods for predicting deep excavation deformation suffer from insufficient accuracy and limited generalization capability. Moreover, the applicability of these methods to different types of monitoring data also requires in-depth analysis. To address this, a machine learning-based prediction model, i.e., the VMD-GWO-CNN model, integrating Variational Mode Decomposition (VMD), the Grey Wolf Optimizer (GWO), and the Convolutional Neural Network (CNN), is proposed to predict various types of monitoring data. The GWO algorithm optimizes both the key parameters of VMD and the hyperparameters of the CNN. The optimized CNN model predicts each subsequence decomposed by VMD, and the final prediction is obtained by superimposing these results. Furthermore, the prediction performance of the proposed model is evaluated against the LSTM, CNN, and GWO-CNN models using four metrics (RMSE, MAE, MAPE, R²). The results indicate that all four algorithms possess effective predictive capability for the monitoring data, in which the VMD-GWO-CNN model demonstrates the best performance across all metrics. Specifically, its RMSE for surface settlement prediction is reduced by 59.2%, 34.1%, and 33.0% compared to the LSTM, CNN, and GWO-CNN models, respectively. Moreover, the VMD-GWO-CNN model exhibits strong predictive performance for deformation in slope engineering and subgrade engineering, demonstrating its good applicability across different geotechnical engineering. The findings provide a scientific basis for safe excavation construction and contribute to efficient and rapid execution of foundation pit projects.