Hydromechanical evaluation of flow behavior in rock foundations of concrete gravity dams

This thesis presents a methodology to evaluate the hydromechanical behavior of concrete gravity dams founded on jointed rock. The purpose of the study was to examine the assumptions on current design guidelines (FERC, USBR) and to propose a methodology to extrapolate dam behavior to unusual loading (PMF) conditions. The research methodology consisted in creating a database of observed dam behavior throughout typical cycles of reservoir filling and simulating this behavior with a coupled hydromechanical numerical model.; A hydromechanical model using UDEC 3.0, a computer code based on the Distinct Element Method (DEM), was implemented to fit site specific conditions for Albigna and Hungry Horse dams. However, in Upper Stillwater dam, high effective normal stresses at the foundation precluded any significant hydromechanical interaction. Therefore, the Upper Stillwater case history was not further evaluated with the DEM hydromechanical model; instead, its monitored performance was interpreted on a qualitative basis.; The main concluding points of the study can be summarized as follows: (1) Present design uses a constant cohesion intercept along potential sliding surfaces, regardless of the magnitude of the effective normal stresses, to explain satisfactory structural behavior observed in concrete dams. The results obtained in this study indicate that adequate dam behavior might actually be the result of factors other than cohesion, such as the presence of 3-D rock wedges at the dam foundations, the arching of reservoir water loads into the abutments, the hydromechanical interaction along foundation joints and enhanced natural drainage. (2) For the dams analyzed in this study, estimated changes in rock joint permeability within dam foundations during reservoir filling are in the range of one to two orders of magnitude along the dam rock interface instead of the extreme permeability contrast assumed in current design guidelines (FERC, USBR), which assume open crack at full reservoir and closed at the tip of the crack. (3) The hydromechanical analyses carried out in this research indicate a gradual reduction in factors of safety during reservoir filling assuming that drains work beyond maximum operating elevations, instead of the rapid reduction obtained with design guidelines.