Hybrid Simulation of the Seismic Response of Squat Reinforced Concrete Walls - Wall 2 Test

Reinforced concrete structural walls are commonly used in industrial and nuclear facilities as primary seismic lateral-force-resisting components. However, there is significant uncertainty about expected strengths, deformation capacities, and failure modes of these walls in earthquake load sequences, such as main-shock/aftershock combinations. Hybrid simulation is an effective experimental method to examine these issues because it enables simulation of the seismic response of squat and thick shear walls without the need to recreate the often very large mass associated with the remainder of the prototype structure. A hybrid simulation program at the nees@Berkeley laboratory utilized two shear wall specimens to investigate the variations of squat wall responses caused by different earthquake magnitude sequences.

Conventional displacement-controlled hybrid simulation methods are not appropriate for stiff specimens since small displacement increments correspond to large force increments, and even small displacement errors can lead to significant simulation errors, and possibly to instability of the simulation setup. In order to ensure a smooth force response, a displacement-controlled hybrid simulation setup was made using a displacement encoder with a resolution of 10 microns in the displacement feedback loop. The numerical portion of the hybrid simulation of seismic response was an added mass, modeled using OpenSees and OpenFresco. The stiffness of the wall was estimated using an extremely small deformation triangle wave stiffness test. The mass was adjusted to match the period of a typical nuclear structure, selected at 0.14 seconds based on a Candu reactor prototype. The 1999 Kocaeli, Turkey ground motion was matched to the Diablo Canyon Nuclear Power Plant design basis earthquake (DBE) spectrum and used as the base motion for the hybrid simulations. This motion was further scaled to attain specific target behaviors in three levels of motion: an operational basis earthquake (OBE), a design basis earthquake (DBE), and a Beyond Design Basis Earthquake (BDBE). Since nuclear facility structures are designed to remain “essentially elastic” in a DBE, they should remain elastic or only yield marginally in that event. Thus, the OBE motion was designed so the wall would reach about 1/3 of its yield force. The DBE level motion targeted the wall reaching about 2/3 of its yield force. Then the BDBE was scaled to be an extremely large event that would enable investigation of the post-peak-strength behavior of the walls. This was selected to be 3 times larger than the DBE. Wall 1 was tested with a sequence of increasing levels of ground motion (OBE-DBE-BDBE), followed by a DBE-level aftershock. Wall 2 was tested with an OBE (assuming it had been in service) and then a large BDBE ground motion, followed by two DBE-level aftershocks. Following ground motion sequences, both walls experienced single cycles to 1 in and a cycle to 1.5 in.

The nominally identical walls were 8 in thick models of a prototype 36 in thick structural wall found in a typical nuclear power plant structure. The walls were 10 ft long and 5 ft, 4-1/8 in tall to the height of the actuator axis (aspect ratio 0.53). They had a 0.67% horizontal and vertical wall reinforcement ratios with #4 reinforcing bars placed in two curtains at 7 in on center. The concrete mix design had a target compressive strength of 5000 psi.

Both walls initially experienced shear and then flexural deformations. As the displacement demands grew, the flexural cracks from each end of the wall joined to form a continuous crack along the interface between the wall and its foundation. Sliding in the narrow zone surrounding this crack initiated and grew to dominate the response of the walls. The eventual failure mode of both tests specimens was sliding shear. The force-drift responses of both walls were similar, indicating that significantly different sequences of ground motion intensities do not result in a significantly different force-deformation response of squat shear walls nor do they affect the failure mode sequence for these walls. Wall 2 achieved slightly higher force levels than Wall 1 due to its relatively smaller amount of damage when it experienced the largest BDBE ground motion.

Please note that the Derived_Data "*.d1m.txt" (and subsampled "*.d5m.txt" or "*.d10m.txt") files are identical to the Converted_Data "*.nees.txt" files, but the headers are reduced to a single line for convenient processing. The subsampled files are provided for faster data processing times.