Seismic Response Analysis And Parameter Study Of Tunnel Under Different Buried Depth

Tunnel has become one of the important means of transportation,and more and more attention has been paid to the seismic research of tunnel.The buried depth of tunnels has an important effect on the dynamic response of tunnels.At present,the seismic response analysis of underground tunnels under different buried depths is not perfect enough.Most studies only consider the impact of seismic waves under vertical incidence,which is deviated from the actual engineering.At the same time,the influence analysis of different parameters under different burial depths and the application of machine learning algorithm in underground engineering earthquake resistance are also less.On this basis,the seismic response analysis and parameter study of tunnels under different buried depths are carried out.By means of wave theory and finite element software ABAQUS,this paper studies the seismic response of typical tunnels under different buried depths,and analyzes the influence of ground motion incident Angle,incident wave type,incident wave type,tunnel cross section,tunnel diameter,single and double lines and other parameters.Decision tree algorithm,neural network algorithm and SVM algorithm are introduced to analyze the influence and sensitivity of various parameters in the process of tunnel seismic response.The main work of this study is as follows:(1)The seismic response analysis of typical tunnels under different burial depths was completed,and the dynamic response rule of tunnels under different burial depths under oblique incidence was grasped.Based on the equivalent nodal force of viscoelastic boundary and programmed input,set up the analysis model,typical tunnel under the conditions of different buried depth of tunnel under different buried depth in seismic response analysis shows that the tunnel under perpendicular to the axis and parallel to the axis of the incident response rule,namely in the oblique incidence of the perpendicular to the axis,the maximum stress appears in inverted arch and vault tunnel,With the increase of buried depth,the peak stress of tunnel decreases gradually.In the oblique incident parallel to the axis,the longitudinal stress is greater than the circumferential stress at the same buried depth,and the longitudinal stress is uniformly distributed in the section and decreases with the increase of the buried depth.(2)Based on the verified soil structure action analysis method,the influence law of each parameter in tunnel response analysis under different buried depths is studied and grasped.The variation law of seismic response is studied by changing the parameters.The results show that the tunnel circum-tensile stress under the action of SV wave plays a key role in the tunnel failure under the condition of different tunnel buried depth.When the buried depth is within 15 m,the circumferential tensile stress increases with the increase of incident Angle.Horseshoe tunnel and large diameter tunnel are not conducive to earthquake protection,and the damage is easy to occur in the discontinuous curvature of the tunnel.Seismic waves with different spectral characteristics,even if the peak acceleration is the same,will cause different degrees of dynamic response.Under the same seismic wave,the failure location of the double-line tunnel is the same as that of the single-line tunnel,but the peak stress of the double-line tunnel is greater than that of the single-line tunnel.(3)Based on the neural network regression algorithm,the tunnel stress prediction model was established,and the sensitivity of each parameter in the dynamic response of the tunnel was studied.By comparing the prediction accuracy of decision tree algorithm,neural network algorithm and SVM algorithm in the tunnel seismic analysis,the neural network algorithm is selected as the basis of the prediction model,on this basis,the stress prediction model is established,and the order of the influence of each parameter on the dynamic response is obtained by using the control variable method.