In past earthquakes there have been a number of reports on damage to liquid storage tanks. The earthquake damage has been due to a number of causes with the most common being buckling of tank walls and uplift at the tank anchorages. There has been a significant amount of research activities over the past several decades on the seismic analysis and design of liquid storage tanks. These research activities have been devoted to improving the seismic performance of liquid tanks. The seismic behavior of liquid tanks, however, are quite complicated due to the liquid-structure interaction. An alternative approach to increase the seismic resistance of a liquid tank is to reduce the seismic forces, to which a tank is subjected, by incorporating base isolators. The role of base isolators under seismic loading is to isolate the structure from the horizontal components of the earthquake ground movement, while the vertical components are transmitted through to the structure relatively unchanged. Base isolators, mainly employed to isolate large structures under earthquake ground excitations and rehabilitate structures damaged by past earthquakes, both deflect and absorb the seismic input energy horizontally transmitted to the structure. Design methods of base isolation systems have been improved and experimental studies on base isolators have been performed by many investigators. When the inelastic properties of base isolators are defined, it is usual to model the base isolators by simple models, such as a bi-linear model, in numerical analysis procedures. However, it would be desirable to conduct experimental studies on structural systems with base isolators since inelastic behavior of a base isolation system is too complicated to be represented by a simple numerical model. Hence, the seismic performance of base-isolated structures has been verified experimentally by shaking table tests. However, it may be difficult to perform the full-scale shaking table test of base-isolated structures. In many cases, base-isolated liquid storage tanks such as oil tanks and LNG tanks are too large to perform a shaking table test in full scale. In this study, a pseudo-dynamic test method for evaluation of seismic performance of base-isolated liquid storage tanks is introduced. The pseudo-dynamic technique is to simulate the seismic response quasi-statically in a laboratory by means of on-line computer controlled experiment. But the liquid-structure interaction of tanks can hardly be simulated by the pseudo-dynamic test conducted in a quasi-static manner. Thus, in this study, the pseudo-dynamic technique applicable to base-isolated liquid storage tanks was developed by incorporating a substructuring technique. The pseudo-dynamic test method incorporating a substructuring technique has been mostly developed and applied by Dermitzakis et al. and M. Nakashima. Using a substructuring technique, a base-isolated liquid storage tank is divided into two parts in the pseudo-dynamic testing algorithm. The base isolators are tested as the tested part, and the liquid storage tank is treated in a computer with an appropriate modeling as the computed part. The liquid storage tank was modeled as a discrete system based on the simplified mechanical model derived by Haroun. In the present study, major interest was placed on the development of a pseudo-dynamic test method incorporating a substructuring technique to simulate the hydrodynamic behavior which can not be accounted for in a quasi-static manner. Two types of liquid storage tanks were tested to predict the seismic loads exerted by the liquid on the tank wall in terms of hydrodynamic forces and the effect of base isolation systems. The pseudo-dynamic test method proposed in this study can be applied to the seismic response prediction of base-isolated liquid storage tanks on quasi-static testing facilities.
