The present work relates to the study of finite, linearly elastic cylindrical shells, subjected to earthquake motions. The mechanism of earthquakes is not fully understood. The occurrence, location, and intensity of earthquakes cannot be predicted with certainty. Earthquake Engineering therefore is not an exact science...
...This thesis shows that: (1) By replacing the half-space with a set of elastic springs, the analysis is restricted to the case of a lumped mass system, for the soil-structure formulation. (2) When the half-space is replaced by a finite elastic-continuum the dynamic response of a cylindrical shell is calculated by formulating the Lagrangean of a dynamical system, in which the interaction forces between the soil and the foundation are fully determined and form an integral part of the Lagrangean formulation. The computer programs developed are in accordance with the type of analysis followed. Differences are noted and analyzed...
...based on the results obtained in the study of the Transfer Function Characteristics and the ensuing earthquake response of our model the following general conclusions can be drawn:
1) The free-field accelerations are generally attenuated at the base of a building that is founded on a compliant soil. The amount of attenuation increases when the value of the Shear-wave VS decreases.
2) The natural frequency of the fundamental mode for the models decreases when VS decreases.
3) The motion associated with the mode of vibration, presents structural distorsions which are directly related to the soil stiffness.
4) Lateral displacement effects at the base of the model cannot be determined when considering lumped mass models.
5) Rocking contributes overwhelmingly to the total response of the Shell models when founded on soft soils, and less significantly on stiffer soils.
6) For the case of the Shear-wave, i.e. VS = 1700. ft./sec, the dynamic analysis of a lumped mass model, will yield reasonably good but slightly conservative displacements.
7) If high frequency Transient loads, are considered in the boundary value problem, the Fourier-Laplace Transform Method should be used.
With respect to possible future research work the following comments are offered:
(a) The Influence Functions for higher values of the wave numbers kβ should be calculated, if possible in p conjunction with the Transient problem.
(b) Comparative effects for the P-wave(Pressure-wave) and the S-wave(Shear-wave), should be analyzed independently; that is a more selective analysis may be carried out.
(c) Short-thick, long-thick and long thin cylinders might be considered in order to determine the effects of length and thickness parameters.
(d) Other shell geometries will not be fruitful and are not recommended.
It is evident that it is necessary to establish a different approach for an examination of the postyield behavior for the Cylindrical Shell.