基于隧道施工诱发地表沉降随机介质理论预测模型的拓展

Based on extension of stochastic medium theory prediction model for surface subsidence induced by tunnel construction

为了对隧道开挖过程中产生的地表位移变形大小有一个准确合理的预估,基于随机介质理论法推导了隧道不同断面形式下的非均匀收敛预测模型。利用实测沉降值,采用并行退火遗传算法对模型中主要参数识别的复杂非线性问题进行优化。依托西安地铁八号线区间隧道(电子城站—东仪路站)工程案例,对马蹄形隧道的预测模型进行验证,并采用MIDAS/GTSNX建立有限元模型,对提出的模型预测结果与数值计算值进行对比;通过穆陵关隧道左线进口地表实际监测值对圆形隧道的预测模型进行验证。结果表明:应用提出的预测模型时,该地区地质条件下马蹄形断面隧道中顶部收敛值与底部收敛值分别取26.14、7.28 mm,穆陵关隧道左线进口地表沉降预测结果与实际规律具有良好的一致性。

In order to have an accurate and reasonable prediction of the surface displacement and deformation generated in the process of tunnel excavation,the non-uniform convergence prediction model under different cross-section forms of tunnel was derived based on the random medium theory.The measured settlement value,the parallel annealing genetic algorithm was used to optimize the complex nonlinear problem of the main parameter’s identification in the model.The prediction model of horseshoe tunnel was verified based on the tunnel project between E-city Station to Dongyi Road Station of Xi’an Metro Line 8,the finite element model was established by MIDAS/GTSNX,and the prediction results of the model and the numerical calculation values were compared.The prediction model of circular tunnel was verified by the actual monitoring value of the entrance surface of the left line of the Mulingguan Tunnel.The results show that when the model in this paper is applied,the top convergence value and bottom convergence value of the horseshoe-shaped cross-section tunnel in the geological conditions of the region are 26.14 and 7.28 mm,respectively.The prediction model of the entrance surface of the left line of the Mulingguan Tunnel is in good agreement with the actual law.5 tabs,7 figs,30 refs.