VIBRATION ANALYSIS OF DAMS

Abstract

In order to simplify the aseismic design of dams, it is conventional to replace the dynamic forces by equivalent static forces and these are usually given as a function of the weight of the structure. This function depends upon the seismicity of the site, mass and stiffness distribution and damping of the structure and is generally known as seismic coefficient. In actual practice, the design seismic coefficient is adopted keeping in view the seismicity of the site but without giving any regard to the dynamic properties of the structure itself. The codes of practice of various countries specify different values of seismic coefficients for the design of dams but these appear to be essentially empirical and have no rational justification. The seismic coefficient method may be reasonable for preliminary design purposes but will generally lead to inefficient final designs. In order to arrive at an effi cient design, adetailed dynamic analysis of the dams is essential. This thesis deals with the dynamic analysis of dams subjected to earthquake excitation. Gravity dams, both concrete as well as earth have been analysed. The analysis has been carried out considering elastic behaviour of the dam material. Two dimensional behaviour ofr t+huea dAa*mms* hnaass bDeeeenn considered. Finite element method has been useda iinM t+hKoe sannaallvyssiiss.. One dimensional analysis has also been carried out at aopropriate places for comparison „lth two dimensional Saviour. Dynamic response has been evalua te superposition method. The ground motion data of ted using mode superposltl two typioal accelerograms, namely, Koyna a I- evaluating dynamic response. Extensive (1940) has been used in evaluating Ill use of digital computer has been made in this investigation. An efficient digital computer program has been developed for this purpose. The analysis gives dynamic displacements, accelerations, static stresses, dynamic stresses and total stresses throughout the dam cross-section. Two points investigated are the effect of the vertical component of ground motion and the influence of reser voir water. The investigation shows that in the finite element method of analysis, a dam can have both horizontal and vertical motions even under horizontal excitation alone unlike in the one dimensional analysis where horizontal excitation will cause hori zontal motions only. When the analysis is based on the finite element method, the contribution of the vertical component of ground motion on the dynamic response is small. The additional forces due to reservoir water when compared with the hydrodynamic pressures may be obtained if the hydrodynamic pressures are eva luated corresponding to the peak ground acceleration of an earth quake. The accelerations in the dams during an earthquake vary along the height and width of the dams and the constant acceleration usually assumed for design purposes is not justified. In earth dams, in addition to the effect of the vertical component of ground motion and the influence of reservoir water, the effect of the core on the dynamic behaviour of an earth dam has been studied by considering a central core dam and a sloping core dam. An experimental investigation has been carried out on a '#** IV model of a homogeneous earth dam. The model has been tested on a shake table. The model is also analysed theoretically for the base motion of the table. The accelerations in the model obtained experimentally and analytically have been compared to verify the method of analysis. Bqsed on this investigation, design accelerations for concrete and earth gravity dams have been suggested..