Abstract:
This research comprises of comprehending the current design of Nahakki Tunnel, Swat Twin Tube Tunnel and Shimla Tunnel on the basis of Q, Q and BQ classification systems respectively and analyzing deformations using analytical approaches and Finite Element Modeling (FEM) incorporating PHASE2, to provide a comparison between the analytical and numerical methods, and developing a correlation using neural network. Artificial neural networks (ANN) or connectionist systems are computing systems vaguely inspired by the biological neural networks. During the research work we have faced numerous challenges like deformations at site in Pakistan are either not monitored and if they are, the data is not extensive. Analytical methods show considerable variations among each other, modeling require knowledge and skill in software, an empirical correlation is required that can give accurate deformations effectively. It is found during research that analytical results show similar results like FEM with values lower than FEM. The reasons may be that analytical results don’t take into account the stages of excavation and during analytical calculations the geometry of tunnel is supposed as circular whereas in original the geometry of tunnel is horse shoe.
