Introduction
Currently, spent nuclear fuel (SNF) is stored in onsite independent spent fuel storage installations (ISFSIs), which are dry storage facilities, at 55 nuclear power plant sites. Because the SNF will be stored at ISFSIs for an extended period of time, there is growing concern with regards to the behavior of the SNF within these dry storage systems during earthquakes. To address these concerns, Sandia National Lab (SNL) under the Spent Fuel Waste Disposition (SFWD) program is planning to conduct a series of earthquake shake table tests. The goal of this test program is to determine the strains and accelerations on fuel assembly hardware and cladding during earthquakes of different magnitudes to better quantify the potential damage an earthquake could inflict on spent nuclear fuel rods. Ground motions representative of the different seismic, tectonic and site conditions in United States for seismic hazard levels from 1E-3 to 1E-5 mean annual frequency of exceedance have been developed for this test program to capture the range of seismic environments for dry-cask storage. It is widely known that soil-structure interactions (SSI) effects would alter the ground motions as they interact with the ISFSI concrete pad and dry storage system. These SSI effects should be considered to transfer the ground motions (free field motion of the rock or soil) to the top of the ISFSI pad which will serve as inputs to the Large High-Performance Outdoor Shake Table (LHPOST) at the University of California San Diego (UCSD).
SSI Simulations
The shake table experiments are designed to consist of one dry cask sitting on a concrete pad poured over the platen of the shake table. The effects of the underlaying soil as well as of the neighbor casks as in an actual ISFSI are numerically simulated through SSI analyses and applied as input motion to the shake table.
A set of SSI Analyses are performed with two main objectives: (1) generate (SSI) input motions for the shake table, and (2) simulate the seismic cask behavior to inform the experimental program. A phased implementation is used for seismic simulations.
Phase 1 performs linear equivalent SSI analysis. A model for the SSI system is used consisting of a soil model that represents the site conditions and a structural model that represents the ISFSI: concrete pad and casks. The SSI motions resulting from the linear equivalent analysis, acceleration time histories at the center of the pad, need to be further evaluated, particularly when relative displacements between the cask and the pad occurs, which could invalidate the assumptions for linear equivalent behavior. The output SSI motions are used as input motion for a Phase 2 model. Qualified SSI motions are used as shake table input motions.
Phase 2 performs nonlinear structural analysis. A structural model that represents a concrete pad and a detailed model of one cask, including internals and fuel rods, is used. The interface between the cask and the pad is modeled with nonlinear contact elements, to explicitly simulate the potential relative displacement (sliding and/or uplift) between the cask and the concrete pad. Input motions consist of output motions (including SSI effects) from Phase 1. The site and SSI effects is only implicitly represented through input motions. Seismic behavior of fuel rods is explicitly simulated and evaluated. Critical cases that identify invalid SSI motions are evaluated in Phase 3.
Phase 3 is implemented to perform nonlinear SSI analysis for selected critical cases identified on Phase 2. A model for the SSI system is used which includes an explicit representation of the site and nonlinear structural model of the pad/cask system. The output motions from nonlinear SSI analyses are used as shake table input motions.
Conclusions
The phased SSI implementation allows to generate realistic SSI motions to be used as input motion for the shake table experiments, as well as to evaluate the range of validity for equivalent linear analysis for seismic behavior of dry casks under a wide range of tectonic and site conditions in United States.
The experimental results from the shake table will provide insights about the seismic behavior of spent fuel rods, as well as additional resources to evaluate the performance of the different analysis phases used for this project.