Interaction Between Support And Surrounding Rock Considering Strain-softening Behaviour For Deep Tunnelling

In recent years, a large number of researchers have inverstigated stability of the deep tunnel. Despite of this, tunnel accidents such as the collapse of the rock mass and the broken failure of the support structures occur frequently. The reason for the frequent accidents is that the mechanical response of the deep tunnel exevation, and the rock-support interaction by the tunnel advance are not well-understood by the researchers. Besides, the control measures of the large deformation for the rock mass and the support design are in lack of the theoretical basis and the pertinence. In an aim to overcome the deficiencies, in this thesis, the elasto-plastic analysis of the deep tunnel excavation and the effectiveness of the support by considering the tunnel face effect are investigated systematically and profoundly.Firstly, based on the post-peak mechanical behaviours of rock mass such as elastic-perfectly-plastic, strain-softening and elastic-brittle-plastic behaviour, a finite difference method is adopted to solve the stress and strain distribution in the surrounding rock and the range of the plastic zone. The rationality of the method is validated by the calculation example. The parametric analysis indicates that:for the unstable soft rock with the low strength, the dilatancy parameter and the critical plastic softening parameter (i.e. the plastic shear strain of the rock mass at when the plastic residual stage is arrived) should be highly emphasised. According to the variations of the rock expansion and plastic shear strain versus the confining stress, a nonlinear dilatancy model is proposed. By embeding the model into the proposed finite difference procedure, the nonlinear distributions of the dilatancy angle in the surrounding rock affected by the rock mass quality and support pressure are studied. The result indicates that:for the poor rock mass quality, the dilatancy angle of rock mass near the tunnel boundary is drastically increased, which should be paid attention to; the dilatancy angle decreases but then increases with the growth in the support pressure, and thus, for deep tunnelling and mining, in order to control the dilatancy behaviour of rock mass, the support design can be optimised.Secondly, on the condition that the critical plastic softening parameter is greatly influenced by the confining stress, the unloading path of the confining stress in the plastic zone is given, according to this, a variable confining stress model concerning with the critical plastic softening parameter is developed. A criterion to judge whether the surrounding rock transits from the softening stage to the residual stage is presented. The critical plastic softening parameter and the stress and strain component distributions in the surrounding rock with the constant and variable confining stress model are compared with each other. The result indicates that:for the poor quality rock mass, compared with the variable confining stress model, the constant one tend to underestimate the rock mass deformation to some extent, this condition should be emphasised in the practical engineering. Furthermore, a semi-analytical procedure is presented to derive the stress and strain solutions for deep circular tunnel on the basis of the unified strength theory. The influence of the intermediate stress effect on the rock mass stability is discussed. The result indicates that the support design by the single shear strength theory like Mohr-Coulomb criterion tends to be conservative.Thirdly, based on the two-stage analysis of the rock-support interaction, a simplified numerical procdure to solve the fictitious support pressure with consideration of the elastic-perfectly-plastic, strain softening, elastic-brittle-plastic rock mass is proposed. By comparing with the calculation result by other methods, advatanges of the proposed method are shown. The influences of the critical plastic softening parameter, the rock mass quality, the initial stress, and the dilatancy coefficient on the fictitious support pressure and the tunnel face effect are discussed. The result indicates that:the stress relieve factor in most tunnel modelling simulation or practical tunneling engineering is calculated on the elastic assumption, the corresponding support design is unsafe; in tunnelling with the high initial stress and strong dilatancy rock mass, the unexcavated rock mass far away from tunnel face become unstable early, and thus the pre-reinforcement is recommended.Finally, based on the numerical solution of the fictitious support pressure, the rock-support interaction in the second stage is simulated by the finite difference software FLAC3D. By introducing the rock mass deformation reduction factor, the effect of support on controlling the rock mass deformation is quantitively analysed. According to the parametric analysis, the variations of force in the liner and rock mass deformation reduction factor by different influence factors are discussed. According to this, the specific ranges of the delayed distance of the support installation to the tunnel face and the allowable rock mass deformation are presented. A Modified Ground Reaction Curve is proposed, by which, the relation between the initial rock mass deformation when the support is just installed, and the final support load is presented. The result indicates that for most tunnel cases, the delayed distance of the support installation to the tunnel face should be lower than one radius of the circular tunnel. For the composite support, the effect of the steel set on restricting the rock mass displacement is not obvious. However, it plays a key role in suffering the load imposed on the support, in this aspect, the steel sets tended to'protect'the shotcrete to some extend.