Assessment of Tunnel Stability and Evaluation of Rock Support in Headrace Tunnel of Super Madi Hydroelectric Project, Nepal

Abstract

Underground structure is flexible construction alternative among various other construction. The presence of weathered rocks, difficult slopes and occurrence of frequent hazards in the surface has encouraged the investors and technical personals to work underground for completion of different development related works. Construction of a tunnel, changes in-situ condition and sometime leads to failure if not properly evaluated ground condition. In the most of the case, the risk involved and hazards that may occur is high. Therefore, providing support for tunnel stabilization is important before any failure. Using actual project information of Super Madi Hydroelectric Project the methods to provide tunnel support is discussed in this thesis. Typical site data formed input for the geotechnical engineering design of the tunnel support based on empirical, analytical and finite element modeling. The outcomes of the different approach in the study were unique function of their underlying scientific values. The rock mass was classified using Q value within 400m of headrace tunnel was studied in this research. The Q value varied from minimum of 0.038 at chainage 1+200m to maximum 1.25 at chainage 1+300m. That can be a class of extremely poor rock class to poor rock class. There are three types of rock class within the selected portion i.e. poor, very poor and extremely poor. The finite element analysis was done using generalized Hoek Brown failure criteria for the support estimated from rock mass classification, analytical approach and support used by the project, all these three set of models was tested for different factor of safety. Such support was further analyzed for the block stability and squeezing problem. This thesis is focused to optimize the estimated support for economic project completion. The rock support estimated from the rock mass classification is very first estimation for the different class of rock. These supports are optimized at every section using finite element method. This can conclude that the rock support can be optimized significantly while analyzing individual section with the geological condition instead of generalize the support class for the different category of Q values. Therefore, rock mass classification approach only is not adequate to design and estimation of tunnel support. Numerical analysis is very helpful to estimate the tunnel support in such geological region where rock masses are very poor with high rock cover.