Seismic fluid-structure-soil interaction of buried water reservoirs

Buried water reservoirs are a relatively new class of structures that are increasingly replacing open above-ground reservoirs to gain better control over the water quality and free surface space in addition to many other benefits. Given that many of these lifeline structures are located in seismically active regions, they should be designed to withstand earthquake events. However, evaluating the seismic response of these reservoirs is a challenging problem that involves the concurrent interaction of the structure with the stored fluid and the retained soil, in other words, fluid-structure-soil interaction. The complexity stems from the wide difference in the dynamic response of these individual components when subject to seismic shaking making it difficult to simplify the problem. Despite this fact, the current design practice for evaluating seismic loads and deformations of these structures depends heavily on either design codes and guidelines that are not necessarily applicable or overly simplified non-validated numerical tools. This can lead to underestimation or overestimation of the seismic demands. It is inappropriate to apply design approaches usually utilized for other underground structures to reservoirs because of the reservoir's distinctive layout. Further, the design of reservoirs requires a good understanding of how water reacts to shaking. The water dynamic response can be evaluated by means of analytical, numerical, or experimental approaches. However, it is common in practice to utilize simplified code-based methods that might not address all the mechanisms involved in the dynamic behavior and thus neglect some of the important response features. This shortcoming, along with the computational advancement observed in the last two decades, gave rise to different numerical modeling techniques that can estimate the hydrodynamic behavior of water more accurately. Still, numerical model reliability needs to be validated which is typically done using 1g shake table experiments that exhibit some scaling effects. It is well-known now that water dynamic response is influenced by its interaction with the containing structure, the reservoir structure in this case. Moreover, the reservoir response is highly affected by the seismic loading imposed by the surrounding soil. This implies that decoupling the problem or separately evaluating the seismic response of these individual components would not represent the actual conditions. Therefore, full-fledged numerical models, which include all the components involved, would be the best way to tackle this problem. Such numerical models need to be validated against either field or experimental measurements. To this aim, this study was conducted by combining the results from dynamic centrifuge tests and numerical simulations to evaluate the seismic fluid-structure-soil interaction of buried water reservoirs. The first stage of this study investigated the hydrodynamic behavior of water when subject to shaking while excluding the structure and the soil. The intent was to have a good understanding of the water response before adding more complexity. This would allow numerical models to be evaluated for their capability to predict water behavior, which is perhaps the most difficult aspect of the problem. Also, to the author’s best knowledge, there was no centrifuge experiment available at the time that solely studied the water hydrodynamic behavior. Thus, it was necessary to conduct such an experiment to (1) reduce the scaling effects imposed by the frequently used 1g shaking table and (2) verify its conformity with the centrifuge scaling laws. To this end, a unique series of five centrifuge tests were conducted to investigate the hydrodynamic behavior by varying the height and length of the water body as well as the applied sine wave and earthquake motions. After carrying out the experiments, Arbitrary Lagrangian-Eulerian finite element models were developed to reproduce 1g shake table experiments available in the literature in addition to the centrifuge tests conducted in this study. Also, the experimental and numerical results were compared to the commonly used analytical and simplified code-based methods to highlight their applicability and limitations. The second stage of the study investigated the seismic behavior of buried reservoirs while including fluid-structure-soil interaction through centrifuge experiments and numerical simulations. Two sets of centrifuge tests were carried out at different reservoir orientations in an attempt to investigate 2D motion effects. The centrifuge tests were then reproduced numerically using advanced numerical modeling techniques. The third and final stage included a parametric study to investigate the key factors affecting the behavior of the reservoir. Some of the limitations imposed by the experiments were also relaxed by including more realistic conditions. The study provides practitioners with advanced numerical modeling techniques to better represent reservoirs during analysis and design. It also provides insights into the unique behavior of buried reservoirs and their critical components, leading to a better and more efficient seismic design.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Civil Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2023

Author dc:creator

• AlKhatib, Karim

Contributors dc:contributor

• Hashash, Youssef M. A.
• Andrawes, Bassem O.
• Baser, Tugce
• Olson, Scott M.
• Ziotopoulou, Katerina

• Fluid-structure-soil Interaction
• Buried Reservoir
• Seismic Analysis
• Centrifuge
• Hydrodynamics
• Numerical Modeling
• Finite Element
• Arbitrary Lagrangian-eulerian
• Parametric Study.

Statement dc:rights

• Copyright 2023 Karim AlKhatib

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/120564

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F120564