Full‐scale shaking table test and numerical modeling of a 3000‐liter legged storage tank isolated with a vertical rocking isolation system

Abstract

This paper presents the numerical and experimental evaluation of a vertical‐rocking isolation (VRI). This evaluation is done by 1‐D shaking table tests performed on a full‐scale legged storage tank of 3000‐liters capacity and its representation through a simple yet representative rigid lumped‐mass model approach. The isolation system setup consisted of four ISO3D‐2G devices, each one placed on each leg of the tank, which uses high‐damping natural rubber to generate the restoring and dissipative forces. The ISO3D‐2G device is vertically flexible and laterally rigid, enabling the isolation mechanism of the rocking motion of the tank. The experiments were carried out using three white noise for the system identification and 17 ground motions inputs for the system validation. The measured variables included the lateral acceleration and displacement of the tank, and the vertical and rotational behavior of the isolation interface. The identification results showed a vertical‐rotational coupled fundamental mode that is highly dependent on the amplitude of deformation, with a period varying from 0.5 to more than 1 s, depending on the intensity of the motion. The maximum displacement of the tank at the top remained below 13 cm with total accelerations of nearly 0.3 g, both for motions with Peak Ground Acceleration (PGA) values ranging from 0.3 to 0.8 g. The mean maximum values were predicted with the simplified model with errors of less than 10% and 21% for displacements and accelerations, respectively. Finally, the results show that the behavior of vertical‐rocking isolated structures can be predicted by simplified models with reasonable errors and that the development of simple design guidelines and equations for VRI systems is possible.

Frequency Seismic Response for EEWS Testing on Uniaxial Shaking Table

“…Global RMSEs , computed for all spectral components to be compared with other reported results, have lower values than those reported in recent papers [ 39 , 40 , 41 , 42 ]. Even if simple control algorithms have been used, the minimization of the hardware arrangement, and the low weight of the moving parts (only 390 g for the cart and accelerometer), yields very good seismic-waveform-tracking characteristics.…”

“…Normalization was further included to assess the simulator’s performance, as it is sensitive to the amplitude and frequency errors and provides better identification of the output fitting correctness, similar to reference [ 39 ]. The normalized root mean square error was calculated using the following equation: …”

Seismic Performance Evaluation and Analysis of Vertical Hydrogen Storage Vessels Based on Shaking Table Testing

“…The goal of seismic performance validation for such facilities extends beyond ensuring structural integrity: it also aims to confirm the sustained functionality of systems in the aftermath of an earthquake. While the former can be verified through testing and analytical methods, the latter is commonly known to be efficiently validated through testing procedures [15][16][17]. In March 2019, the Republic of Korea revised its Architectural Seismic Design Standards (KDS 41 17 00) [18], specifying the requirement to substantiate the seismic performance of key non-structural elements for which functionality must be maintained post-earthquake.…”

Effect of ground motion processing and filtering on the response of rocking structures

“…Examples of such systems are the out of plane motion of masonry structures, 12–21 the motion of unanchored equipment, 22–33 and the response of ancient Greco‐Roman and Chinese temples 34–38 . Moreover, allowing a structure to uplift and sustain rocking motion can be used as a seismic design method 39–64 and references therein], because the uplift works as a mechanical fuse and limits the design forces of both the superstructure and the foundation.…”