Research On Nonlinear Prediction Of Large Deformation Grade Of Mountain Railway Tunnel And Supporting Measures

With the rapid development of China's railway system,many unfavorable geological problems in railway tunnels have also received more and more scholars'attention.The construction features of mountain railway tunnels include large excavation sections,long tunnel lengths,high construction risks and high durability requirements for lining structures and so on.For the railway project in southwestern China,due to its complex geological environment and mostly deep buried tunnels,there are many engineering disasters,especially when the tunnel passes through the unfavorable geological area with poor surrounding rock strength.Larger deformations,which in turn lead to damage to the support,can cause tunnel collapse in severe cases,resulting in delays in construction and major casualties and property losses.Therefore,scientific and accurate evaluation of the deformation level of the excavation hole section,and the determination of the support measures according to the deformation level has important research significance.Based on a large number of on-site monitoring data samples,this paper selects three kinds of deformations,namely,arch settlement,upper guide convergence and medium convergence,as the evaluation index,and uses AHP to determine the index weight.The actual deformation of each hole segment in the sample set is graded by extenics.According to the site engineering geological data and relevant construction specifications,the uniaxial compressive strength,rock integrity coefficient,Angle between main structural plane and hole axis,groundwater condition and structural surface state completeness coefficient of rock are selected as conditional attributes,and the deformation level obtained by the extenics method is used as the decision attributes that make up the decision table.The rough set theory is used to analyze the decision table,the weight problem is transformed into the attribute importance problem,the weight value of each influencing factor is calculated,and the weight is compared and verified according to the basic principle of combined weighting.Finally,the normal cloud and the extenics are adopted.The method predicts the deformation level of the tunnel and establishes a prediction model for the deformation level of the railway tunnel based on various nonlinear methods.The engineering application results show that the prediction results of the model are consistent with the actual deformation,which plays a good guiding role for the later tunnel construction.The numerical simulation method was used to study the support measures of different deformation grades.The original support scheme of the tunnel was optimized and a safe and reliable support method was obtained.