Evaluation on the Squeezing Phenomenon at the Headrace Tunnel of Chameliya Hydroelectric Project, Nepal

Growing demand of electricity in Nepal can be fulfilled by hydropower generation. The huge potentiality of hydropower generation in Nepal is mainly due to abundant water resources and available geographical head due to steep Rivers. In medium and large hydropower projects, huge amount of water discharge has to be handled form intake to power station and ultimately back to river again. Also, because of steep topography, the construction of pipe and canal on the surface of terrain could be very difficult and expensive for larger discharges. Hence, underground construction such as tunnels or shafts could only be the feasible options of water conveyance system for large discharges and in case of steep terrains. But, at the same time, there are higher risks and uncertainties associated with the underground works like tunnels and shafts or caverns.The main risks and uncertainties associated with the underground works are stress induced instability, water leakage, mud flows and finally the cost overrun during construction. When there is overstressing of rock mass that means rock stresses exceed the strength of rock mass, there will be stress induced instability in the tunnel. If the rock mass is very weak, schistose and deformable, squeezing phenomenon will occur with the development of plastic zone around the tunnel which causes excessive deformation of tunnel. In the Himalayan region, due to the high degree of schistocity, fracturing and shearing, weak rocks such as mudstone, shale, slate, phyllite, schist, highly schistose gneiss and the rock mass of the tectonic fault zones are not capable to withstand the high stresses. Basically, squeezing has been common phenomenon in the tunnels in these weak and deformable rock masses. In this thesis, Chameliya Hydroelectric Project (CHEP), located in far western region of Nepal, has been taken as the case study. In this project, huge squeezing problem occurred in about 800m stretch of headrace tunnel from chainage 3+100m to 3+900m. The most affected section is about 550m in between these chainages. At several locations in squeezing section, the tunnel wall closure (deformation) has been recorded well over 1.0 m in an average and the maximum above 2.0 m where the original tunnel diameter is 5.2m. Hence, the thesis basically deals with squeezing analysis of the case using different approaches. Rock types along the headrace tunnel alignment are dolomite, slate, talcosic phyllite and dolomite intercalated with phyllite. Mostly, talcosic phylite has been found in the squeezed section. The rock mass quality in the squeezed section is extremely poor to exceptionally poor.The main objectives of this thesis are the assessment of squeezing phenomenon, evaluation of stability of the tunnel and support pressure estimation. In this thesis, four main methods have been used to evaluate the squeezing phenomenon viz.; empirical methods such as Singh et al (1992) and Q-system (Grimstad and Barton, 1993), semi-analytical method such as Hoek and Marinos (2000), analytical method such as Convergence Confinement Method (Carranza-Torres and Fairhurst, 2000) and numerical program Phase2. Initially, seventeen tunnel sections at different chainages have been taken into consideration. The squeezing prediction criteria, such as Singh et al (1992), Q-system and Hoek and Marinos (2000) approach, show that there is severe squeezing in last ten sections. Hence more detail squeezing analysis has been done for these ten sections using Hoek and Marinos (2000) and Convergence Confinement Method, and support pressure has also been estimated using these two approaches and Barton et al. (1974) approach. Hoek and Marinos (2000) and Convergence Confinement Method analysis show that there is significant amount of tunnel deformation to cause squeezing problems. The main factors that control the squeezing phenomenon are the rock mass parameters and rock stresses. Therefore, quality of squeezing analysis largely depends upon the correct estimation of these input parameters. The main components of rock stresses are gravity and tectonic stresses. The rock stresses in the project area were not measured, so Phase2 program has been used to estimate the tectonic stress value from measured deformation. The tectonic stress value has been found to be equal to 3.5MPa in this area, but stress measurement will be necessary to verify this value. Uniaxial unconfined strength of intact rock in four tunnel sections has been back calculated from measured deformations using Phase2 program and found to be in the range of 10 to 15Mpa in the squeezed section. Later, the deformation has been calculated using Hoek and Marinos (2000) and Convergence Confinement Method for improved intact rock strength and compared with Phase2 result. All analyses show that there is significant deformation to cause squeezing problem. In CHEP, tunnel cross section has reduced considerably in several stretches of tunnel. Due to the excessive deformation, temporary supports were provided at several locations, steel ribs and lattice girders are buckled at several locations and shotcrete lining is also cracked. All these have to be removed before application of final lining. Finally, two different possible solutions have been studied using Phase2 program to address the existing problems in squeezed section of the headrace tunnel