The drilling and blasting method exerts a considerable impact on the surrounding rock, causing significant disturbance and vibration, which poses severe stability challenges in weak rock tunnels. To quantitatively evaluate the stability evolution under non-explosive construction compared to traditional methods, this study takes the New Shangjiagou Tunnel of the Xi'an-Yan'an High-speed Railway as a case study. Numerical simulations using LS-DYNA (for blasting dynamics) and FLAC3D (for excavation response) were combined with on-site acoustic wave measurements and deformation monitoring. The results indicate that the non-explosive construction method using a Boom-type tunnel roadheader offers superior stability. Specifically, it reduces the maximum deformation by 5.7 mm (27%) and the plastic zone area by 30% compared to the drilling and blasting method. Furthermore, the roadheader excavation induces a more uniform plastic zone distribution, whereas blasting increases the damage depth to the surrounding rock by approximately 0.5 to 0.7 m. The vertical deformation during roadheader excavation is primarily concentrated in the initial stage, followed by rapid stabilisation. These findings confirm that the Boom-type roadheader is a viable and superior alternative for tunnel construction in weak rock masses, providing a theoretical basis for optimising excavation methods in complex environments.
Thermal integrity profiling (TIP) has emerged as a powerful non-destructive method for evaluating cast-in-place concrete piles by monitoring heat of hydration (HoH) temperature profiles using temperature probes embedded in the piles. However, conventional TIP interpretation relies either on manual inspection of a series of temperature-depth profiles over time, which is inherently prone to subjective bias, or on back-analysis using simplified modelling approaches. This study presents, for the first time, a fully three-dimensional TIP framework combining finite element (FE) simulations with two deep neural networks (NN). A multi-label classifier detects up to three defects per pile, while a regressor quantifies defect volume, location, reinforcement-cage displacement and hydration parameters. A database with hydration temperature profiles from 19,445 simulations encompassing randomised pile lengths, soil stratigraphies, cage misalignments and multi-defect configurations is used to train the two NNs. The classifier achieves a per-defect accuracy of 93.6% (96.8% per sample). The regression model achieves near-perfect accuracy for cage displacement (R2 = 0.992) and defect volume (mean absolute error of 0.03 m3), and an adjusted R2 of 0.92 for defect location. This novel automated approach significantly enhances the objectivity, test speed and reliability of TIP-based defect detection across a wide range of deep pile foundations.
