Face Stability Assessment for Underwater Tunneling Across a Fault Zone

Face stability is critical for the safety of underwater tunnel construction across fault zones. This paper presents a new analytical method that can be used to assess such stability. Following Horn’s model and the double strength reduction method, a new analytical model is proposed which incorporates the effects of groundwater seepage, fault dip angle and excavation footage. By evaluating separately the frictional resistances of the rock and the fault mud of the sliding body, two reduction ratios ($k_1$ for the fault mud and $k_2$ for the rock) are determined. The influences of the fault dip angle, hydraulic pressure, and excavation footage on the tunnel face stability were demonstrated in a case study using the proposed new model. The case study showed that the frictional resistance due to fault mud contributes little to the stability of tunnel face as $k$ increases. With the increase of the fault dip angle, the stability of the tunnel face improves; however, from another perspective, the frictional resistance distributed in the fault mud contributes little to the safety of the whole sliding body as $k$ increases, and the stability of the tunnel face is improved as the dip angle increases, whereas the influence of the fault mud on the whole stability weakens. A smaller fault dip angle causes a more severe warp, indicating that the dip angle significantly affects the shape of the subtraction factor curves. On the other hand, the hydraulic pressure plays a more significant role in determining the shape of the curves compared with the fault dip angle. It was observed that both subtraction factors $f_1$ and $f_2$ were much greater under higher hydraulic pressure. In addition, the overall safety factor $f_s$ decreases with the increase of the distance between the tunnel face and the fault plane ($l$). All curves with different fault dip angles tend to converge to the same safety factor as the value of $l$ approaches zero, whereas different fault dip angles lead to different rates of variation of $f_s$ against $l$. Further numerical results showed that the maximum vertical displacement ($Y_{\max }$) of the tunnel face increases with the reduction of $l$, and the axial displacement of the tunnel face center ($Y_{\mathrm{center}}$) for different fault dip angles is the same when $l$ approaches zero.