Tunnel face instability mechanism in layered ground: Theoretical insights and validation based on numerical investigation and field observation

Abstract

Layered ground represents a prevalent geological formation in shield tunnelling projects, posing significant challenges to the stability of the tunnel face. A critical gap in existing research lies in the lack of a clear distinction between layered ground and specific ground types, such as upper soft-lower hard (US-LH) ground, and the absence of a quantifiable criterion for their differentiation. To address this, the ratio of the unconfined compressive strength (q) of the lower soil layer (q_l) to that of the upper soil layer (q_u) was introduced to delineate soft-hard layer distribution. This study investigates tunnel face instability mechanisms in US-LH ground based on the unconfined compressive strength ratio (q_l/q_u), employing theoretical modeling, numerical simulation and field analysis. The findings were successfully applied to the Xingye Express in Zhuhai, China, yielding the following key conclusions: (1) As q_l/q_u increases, the failure angle (α) progressively diminishes; when q_l/q_u surpasses QR_min (defined as the critical threshold distinguishing layered ground and US-LH ground), the failure zone becomes localized to the upper soft layer, with α stabilizing thereafter. (2) QR_min demonstrates greater sensitivity to variations in internal fiction angle (φ) compared to cohesion (c). Notably, QR_min shows no correlation with increases in C/D or σ_s/γD. (3) It is recommended to prioritize monitoring active instability in the upper soft layer during soil strata excavation, maintaining chamber pressures within 0.8–1.2 times the static soil pressure. Conversely, during rock strata excavation, passive instability in the upper soft layer requires focused attention, with chamber pressures adjusted to 1.3–1.8 times the static soil pressure.