Seismic Response of Tank-Fluid Systems: State of the Art Review and Dynamic Buckling Analysis of a Steel Tank with the Added Mass Method

The thesis is organized into two parts: I. Methods of Analysis and Design of Liquid-Storage Tanks under Earthquake Excitation; II. Numerical Modeling and Dynamic Analyses of a Clamped Steel Tank. Primary objectives are: (a) to provide an overview of the salient aspects of the dynamic response of vertical, cylindrical, ground-supported tanks storing a homogeneous liquid; (b) to present different methods of analysis and design criteria, especially looking at the different codes provisions; (c) to investigate the complex phenomenon of buckling by means of dynamic f.e.m. analyses and by establishing a buckling criterion; (d) to provide comments on the seismic vulnerability of liquid-storage tanks, using the results of dynamic analyses for the development of fragility curves.
The buckling phenomenon plays a fundamental role in steel tanks design due to the small thicknesses used for this class of structures. We investigate dynamic buckling of an aboveground steel tank with an height to diameter ratio (H/D) of 0.40 and with a liquid level of 90% of the height of the cylinder, fully anchored to the foundation and subjected to the horizontal component of real earthquake records. The fluid is modeled using the added mass method, i.e. it is replaced by a system of equivalent masses that exactly returns the hydrodynamic impulsive pressure on the tank walls when the model is accelerated in the horizontal direction. By scaling the accelerograms the study attempts to estimate the critical horizontal peak ground acceleration (Critical PGA), which induces buckling in the cylindrical shell. The finite element model of the tank was subjected to fourteen response-spectrum compatible accelerograms, and dynamic buckling computations (including material and geometric nonlinearities) were carried out using the finite element package ABAQUS. The dominant failure mode was the secondary buckling at the middle-upper part of the tank and the mean Critical PGA was 0.33g, so that this buckling mode should be of great concern to the designer even it is not mentioned by actual codes. It was found that this particular buckling mode is a pure elastic buckling mode but the load level at which occurs is strongly influenced by the occurrence of plasticity in the structure. Then, the results of dynamic buckling analyses were used to assess which is the most efficient ground motion intensity measure and also to build fragility curves. It was found that the most efficient ground motion intensity measure is the pseudo-acceleration at the first natural frequency of the tank-fluid system. But despite this, the fragility curve associated with the PSA has an higher coefficient of variation than the curve in terms of PGA.