ABSTRACT
The integration of geophysical methods in geotechnical engineering offers significant potential in enhancing subsurface characterisation, particularly in tunnelling applications where conventional investigative tools fall short. This study presents the development and application of a novel seismic-based tunnel look-ahead system, termed SmartBoring, which leverages the propagation of seismic waves to detect geological anomalies ahead of tunnel boring machines (TBMs). Tunnelling through heterogeneous ground conditions in urban settings poses substantial challenges, including the risk of encountering unforeseen obstacles, voids, or geological interfaces. These can lead to delays, equipment damage, and safety hazards. By utilizing seismic waves that propagate through the ground and reflect off subsurface discontinuities, SmartBoring enables early hazard detection without requiring cutterhead access or intrusive drilling through tunnel linings. Field demonstrations in two major metro infrastructure projects in Singapore showcase the system's capabilities in identifying subsurface anomalies, including abandoned pile foundations and geological interfaces. The results demonstrate that the system can achieve a detection range of up to 40 meters with a spatial resolution of approximately 2 meters. The study also introduces a probabilistic Bayesian framework to quantify detection confidence, reinforcing SmartBoring's robustness as a forward-looking risk mitigation tool in mechanised tunnelling. This is an expanded version of an earlier paper from the proceedings.
