This study assesses the need for soil-structure interaction (SSI) analysis of surface- or near-surface mounted, seismically isolated nuclear power plants (NPPs). The current rules and guidance for SSI analysis of NPPs are based on the legacy assumption that reactor buildings are stiff and heavy: the two key attributes needed for significant SSI on soil sites. Reactor developers in the United States are considering seismic isolation as a design feature to reduce the impact of the seismic load case and to enable standardization. The substantial reduction in lateral stiffness associated with the introduction of horizontally flexible isolators at the base of a reactor building led to the hypothesis that SSI has no meaningful effect on seismic acceleration and displacement demands on structural components and equipment in surface- or near-surface-founded, base-isolated NPPs. Herein, reactor buildings and their safety-class equipment, the supporting soil domains, and nonlinear isolation systems are explicitly modeled and analyzed to judge whether the hypothesis is correct. An extensive set of response-history analyses was performed for 945 combinations of (1) three fundamentally different, surface mounted reactor buildings, (2) five horizontal isolation systems with a range of linear and bilinear properties, (3) nine seismic inputs covering a range of frequencies and amplitudes of shaking, and (4) seven generic soil profiles that cover a range of sites across the United States and were a part of the Design Certification Documents that enabled the KEPCO APR1400 to be certified for use by the United States Nuclear Regulatory Commission under the 10CFR Part 52 licensing framework. The peak resultant horizontal displacements of the isolation systems, peak resultant horizontal accelerations in each reactor building, and in-structure horizontal and vertical acceleration response spectra were essentially identical with and without considerations of SSI, confirming the hypothesis.

A Fluoride-salt cooled High-temperature Reactor (FHR) that uses circulating solid TRISO pebbles, as a fuel and positively buoyant graphite reflector blocks as a moderator is at an advanced stage of development. Seismic base isolation supports a pathway for cost competitive and rapid deployment of this molten salt reactor (MSR) in regions of different seismic hazard. This paper presents recommendations for analysis and modelling of different components of a base-isolated MSR, based on results of earthquake-simulator tests on a scale model of the reactor. Recommendations are provided for evaluating the sloshing behavior of the coolant in an annular region inside the reactor and for estimating hydrodynamic loads for use in analysis of structural and mechanical components. Hydrodynamic loading on the reflector blocks is discussed and a practical approach to estimate forces in the connections between them is presented. Approaches for modelling and analysis of two 2D horizontal base-isolation systems, utilizing spherical sliding bearings, are validated using results from experiments. The recommendations apply to reactor vessels with similar fluid-structure systems.

The head of a cylindrical reactor vessel must be designed for earthquake-induced sloshing force if there is insufficient freeboard. Herein, a design-oriented calculation method is developed for base-and head-supported annular tanks, where the annulus separates the reactor vessel and its core barrel. Charts are provided to enable preliminary design calculations. Examples are presented to describe the calculations.

One Generation IV nuclear reactor, which uses a fluoride salt as a coolant, graphite reflector blocks as a moderator, and circulating buoyant TRISO pebbles as fuel is at an advanced stage of development. To characterize the seismic behavior of components of this reactor, validate numerical models for analysis, and develop recommendations for design, a set of earthquake-simulator experiments on a scaled model of the reactor vessel and its internals was executed on a six-degree-of-freedom earthquake simulator. The model was seismically isolated at its base using two types of spherical sliding bearings. The scaled model involved representations of the prototype reactor vessel, core barrel, reflector blocks, coolant, and spherical fuel pebbles. The material and geometric properties of different test components were selected based on a dynamic similitude scaling analysis and an approximate length scale of 0.4. Four sets of three-component earthquake motions were used as inputs for testing. Instrumentation on the test specimen recorded the dynamic responses of the outer vessel, core barrel, and reflector-block assembly, the hydrodynamic responses (sloshing and hydrodynamic pressure) of the liquid coolant, pebble consolidation under earthquake shaking, and the behavior of the isolation systems. This paper describes the design of the experiments and presents key results from the tests. The dynamic responses of the outer vessel, core barrel, and the reflector blocks revealed that the components responded as a unit for the intense shaking used in the experiments. The sloshing response of the fluid in a thin annulus near the perimeter of the vessel was heavily damped. The change in the packing fraction of the pebble bed under repeated, intense 3D earthquake shaking was less than 3%. Seismically isolating the vessel substantially reduced demands on its internal components.

Earthquake shaking more intense than that used to size the horizontal clearance between a base-isolated building and near-rigid perimeter moat wall will result in hard impact, producing high-frequency, high-amplitude acceleration response in the structure and supported equipment. This paper provides a design solution for the damaging effects of hard impact by installing a compliant engineered element in the load path between the base-isolated building and the moat wall, resulting in soft impact and a much smaller acceleration response. The engineered element assumed herein is a commercial-off-the-shelf marine fender with mechanical properties determined by physical testing. The attachment of a flexible engineered element, with well-defined stiffness and damping, to a near-rigid moat wall, simplifies the numerical modeling of the building-moat wall system and eliminates the need to bound the lateral stiffness of the wall for impact calculations. The simple model of the engineered element can be implemented in commercial finite element codes. Theory is developed for two-sided impact of a single-degree-of-freedom oscillator. Analytical solutions are derived for the shifted first-mode frequency of the impacted oscillator and for its free-vibration response. The shifted first-mode frequency is a function of the composite lateral stiffness of the isolator-engineered element assembly and its earthquake-induced displacement. Local peaks in the spectral response of the impacted oscillator form at odd integer multiples of the shifted first-mode frequency. The analytical solutions can be used to verify, in part, the numerical model used for impact analysis.

The seismic response of safety-related equipment mounted on the head of an advanced reactor, including pumps, control rod drive mechanisms, and reactor monitoring devices, will affect the design and layout of many advanced reactors. High earthquake-induced accelerations in such equipment may challenge their seismic qualification and trigger the need for additional support framing on the reactor head. Base isolation is a design solution that can drastically reduce seismic demands on equipment. This article describes a set of earthquake-simulator experiments conducted on a scale-model of a base-isolated reactor vessel including four representations of head-mounted equipment, with frequencies spanning from 4.5 to 27 Hz. Dynamic responses of the head-mounted equipment, including displacements, accelerations, and strains, were measured in the experiments for three support conditions: conventional, and seismically isolated using single concave Friction Pendulum (SFP) bearings and triple Friction Pendulum (TFP) bearings. Seismic isolation was effective at reducing equipment responses (accelerations, displacements, and strains) with respect to those in the conventionally supported vessel across a range of seismic inputs. Companion numerical studies highlight the accuracy to be expected in the calculation of different response quantities for lightly damped equipment. The importance of characterizing damping in head-mounted, safety-related equipment through physical experiments to support design and risk assessment is made clear through the numerical simulations.

Standardizing advanced nuclear reactors is a pathway to substantially reducing their overnight capital cost and achieving parity with other power sources, including renewables and fossil fuels. The seismic load case has thwarted standardization of nuclear power plants because site-specific seismic hazard and local near-surface geology has triggered soil-structure-interaction analysis, design, equipment qualification, regulatory review, and licensing, ensuring that each build is different. To achieve standardized or site-independent certified advanced reactor designs, the impact of the seismic load case on the engineering and construction cost and time must be substantially mitigated. Seismic isolation is a mature technology that has been used for more than 30 years in non-nuclear sectors to substantially reduce earthquake demands in buildings and other infrastructure. In this paper, seismic isolation is used to enable standardization of advanced reactor designs, aimed at the complete re-use of a site-independent, certified design and repeated procurement of safety-class equipment. A pathway to standardized designs using seismic isolation is demonstrated for two fundamentally different advanced reactors: a molten salt reactor and a high temperature gas reactor. Each reactor building is equipped with three specialized pieces of safety-class equipment, namely, a reactor vessel, a steam generator, and a control rod drive mechanism housing that is attached to the reactor head. Analysis is performed per ASCE and ASME standards to design the buildings and the equipment for two base conditions: conventional (fixed base) and base isolated. The impact of the seismic load case is characterized for the reinforced concrete walls in the buildings and for the equipment, measured using vessel wall thickness and horizontal accelerations. The analysis results show that the fixed-base buildings, designed for a site of low seismic hazard (peak ground acceleration, PGA = 0.15 g) could be constructed at a site of much greater seismic hazard (PGA = 0.7 g) if seismic base isolation is employed. Importantly, the scope of the site-specific analysis, design, and qualification would be limited to the seismic isolators and the isolated substructure, drastically reducing plant-specific engineering, review, and licensing, and time to construction start. Regulatory challenges and opportunities with standardized reactor designs are identified.

Seismic analysis of structures, systems, and components (SSCs), including the consideration of soil-structure interaction (SSI) effects, is an important and required step in the design and licensing of nuclear power plant SSCs important to safety. Historically, the SSI analysis of nuclear structures has been performed using equivalent linear methods. However, there has been considerable industry investment in alternative seismic design and analysis approaches to reduce the construction cost of new reactors. Toward that goal and in alignment with the Licensing Modernization Project (LMP) framework, the Nuclear Regulatory Commission (US NRC) has proposed a risk-informed, performance-based approach to seismic design that allows inelastic response in those nuclear plant structures not required for confinement. As a complementary effort, reactor designers are exploring nonlinear seismic analysis methods to optimize structural designs and reduce construction costs.

Nuclear energy has a key role to play in global decarbonization. Impediments to the widespread deployment of reactors are their projected high capital cost and levelized cost of energy, and time required to analyze, design, license, construct, and commission them. The earthquake load case is a key cost driver for a new build nuclear plant, because near-surface soils and seismic hazard are different at each site, requiring site-specific analysis, design, engineering, qualification, licensing, and regulatory review, essentially making every design First-of-a-Kind (FoaK). To enable deployment at the scale needed for deep decarbonization, the cost and time impact of the seismic load case must be significantly mitigated, and plants must be standardized. Seismic base isolation has been proven to considerably reduce the earthquake response of structures and equipment but has yet to be applied to a nuclear power plant in the United States, in part because the financial impacts, positive or negative, are not known. Because there are no recent non-proprietary data to characterize the influence of the seismic load case on capital cost, it is difficult to confidently quantify the financial benefits of seismic isolation.& nbsp;Scheme-level designs of two fundamentally different advanced reactor buildings were developed to assemble cost data on the influence of the seismic load case. Both buildings were equipped with three bespoke pieces of safety-related equipment and analyzed for incremented levels of earthquake shaking to quantify the seismic penalty on equipment, in terms of vessel weights and horizontal accelerations. Using analysis results, a questionnaire was developed and transmitted to nuclear utilities, reactor developers, engineers, and equipment suppliers to collect cost data on engineering and fabrication costs for these unique pieces of safety class equipment. Synthesis of the cost data showed that the seismic load case significantly affects the capital cost (sum of engineering and fabrication cost) of safety class equipment, with engineering costs being comparable to fabrication costs. Standardization of safety-class equipment is made possible by seismic isolation, that is, equipment designed for minimal seismic robustness can resist earthquake shaking at a site of much higher seismic hazard. The average reduction in the capital cost of the safety-related equipment, enabled by seismic isolation, is a factor of two for FoaK equipment and a factor of five for standardized equipment.

The benefits of seismically isolating nuclear power plant buildings, in terms of reducing seismic risk, are well established but the possible impacts on overnight capital cost are unknown. Projects funded by EPRI and ARPA-E are now underway to characterize possible reductions in overnight capital cost of new build plants, with a focus on the financial impact of the seismic load case. The EPRI-funded study is addressing the base isolation of reactor buildings and the ARPA-E MEITNER project is assessing the use of equipment-based seismic protective systems in advanced reactors.Two generic reactor buildings were designed to provide data on equipment weights and lateral accelerations as a function of incremented levels of earthquake shaking. One building houses a molten chloride fast reactor and the other a high-temperature gas reactor. Each building was populated with three pieces of equipment: a reactor vessel, a steam generator and a housing for a control rod drive mechanism. Response-history analysis was performed using earthquake ground motions consistent with the seismic hazard at the Idaho National Laboratory site, in Idaho Falls, ID. The minimum required wall thicknesses for the reactor vessels and steam generators in these buildings, for operational and incremented earthquake loads, are reported. Lateral accelerations, which are used for the seismic design of the internals in these vessels and for the control rod drive mechanisms, are presented for incremented peak ground shaking of the buildings. The benefits of base isolating the two buildings, in terms of reduced thickness of vessel walls and horizontal accelerations of …

The seismic SSSI effects in the nuclear island buildings for a representative plant are considered. The nuclear island consists of four buildings. The seismic SSSI effects on the lightest and smallest building (building of interest) are analyzed with two different approaches (ASCE 4-16, 2017). First, a model that includes all four buildings (explicit SSSI model) is constructed and used for the SSI analysis under a reference ground motion. This represents a rigorous approach that streamlines the steps necessary, but increases the complexity and size of the analysis model. However, taking advantage of high performance computing capabilities (HPC), the time and effort of the analysis remain within reasonable limits. Second, a two-step procedure is used for an approximate study of the SSSI effects on the building of interest: (1) a model that excludes the building of interest, but includes all other three buildings within the nuclear island (neighbor buildings model) is used for a seismic analysis with a reference ground motion to calculate the seismic response as acceleration time histories at various points on the soil at the footprint of the building of interest, which are used to create a modified input motion that includes the effects of the nearby buildings; and (2) the modified input motion is used for a seismic analysis of a stand-alone SSI model of the building of interest. This represents a simplified approach, which is computationally less intensive, but restricts the interaction only to the input motion and neglects to consider the mutual interaction between buildings during their seismic response. The SSSI effects are evaluated comparing the seismic response of the lightest and smallest building (building of interest) from the explicit SSSI model with the response of the same building of interest without SSSI effects, using a model with no other buildings included (stand-alone SSI model) both under a reference ground motion. The structural response from the rigorous approach is compared with the response from the simplified approach to assess the effectiveness and limitations of the simplified approach. It is found that the response spectra results from the simplified approach significantly overestimates the peak response amplitude compared to the response spectra results from the rigorous approach.

This paper introduces a new alternative for soil-pile-structure interaction analysis, based on established soil-pile interaction theory, which accommodates a wide range of foundation conditions. A description of the theoretical formulation and its implementation into the prevalent SASSI software framework via SCSASSI (SC Solutions, 2018) is provided in this paper, including examples and usage recommendations. A companion paper (Rangelow et al, 2018) discusses benchmark results (impedance functions) of this implementation derived for various pile foundations. A key component to solve the soil-structure interaction problems in the frequency domain (such as in the SASSI framework) is the compliance matrix. The traditional formulation for the compliance matrix implemented in SASSI (Lysmer et al., 1981) has proven to be effective for the simulation of shallow foundations, but it is found inadequate for simulation of pile foundations. An alternative formulation for the simulation of pile foundations builds the compliance matrix for the portion of the soil where piles are embedded using Green’s function formulations for ring and disk loads, which are semi-analytical formulations to calculate displacements both outside and inside of a cylindrical core. The Green’s functions are evaluated using the equivalent radius of the pile, considering a disk load distribution at the interaction node located at the pile tip, while ring load distributions are considered at the remaining interaction nodes along the pile shaft. The Green’s functions approach for computation of the compliance matrix for pile foundations is implemented in SC-SASSI (SC Solutions, 2018), together with pile finite elements (similar to mechanical formulation of beam elements, but accounting for the effect of the excavated soil) to permit the simulation of 3D soil-pile-structure interaction. The implementation has been generalized to supplement (rather than replace) functionality of the traditional SASSI framework via interaction node groups, where the compliance matrix can be calculated using Green’s functions or traditional formulation for different regions of a foundation model corresponding to deep (pile) or shallow foundations, respectively. This implementation allows for analysis of deep foundations, shallow foundations, or a combination thereof in a single model, in addition to modeling single piles, pile groups, and pile configurations with varying pile diameters. Verification and sensitivity analyses using the SC-SASSI pile element implementation have been performed to demonstrate the technical basis, modeling flexibility, best practices, and recommendations for usage. 25 Conference on Structural Mechanics in Reactor Technology Charlotte, NC, USA, August 4-9, 2019 Division III

A comparative seismic soil-structure-interaction (SSI) study was performed for the reactor building of a nuclear power plant in Switzerland in order to gain insights into the relative significance of nonlinear response effects which may be triggered during beyond-design-basis seismic events. The case study reactor building is founded on 20m of gravel underlain by bedrock and with an embedment depth of 9m.The analysis program was designed as a sequential study to investigate both the independent and combined effects of nonlinear response of soil, soil-structure interface, and structure on the characteristics of in-structure response spectra (ISRS) throughout the reactor building at a hazard level greater than the design basis earthquake. An integrated finite element (FE) model of the soil-structure system, including soil continuum and detailed structure, was built to perform nonlinear time domain analysis (NLTD). For comparison to the commonly employed equivalent linear approach in frequency domain (ELFD), the modeling assumptions in time domain were made consistent with the ELFD approach in an equivalent linear time domain analysis (ELTD). Subsequently, the nonlinear effects were added to the ELTD model, one at a time, to create various NLTD models. Finally, different combinations of nonlinear effects were considered. The interface nonlinearity was introduced through contact surfaces. Soil nonlinearity was incorporated through a hysteretic plasticity model whose shear response is dependent on soil effective pressure. To calibrate the plasticity model, gravel’s shear stiffness degradation curve was modified to produce shear strength values consistent with the laboratory-measured friction angle. Composite layered shell finite elements with nonlinear material properties were used to model key structural shear walls. The reasonable match between ELFD and ELTD results confirmed the general viability of time domain approaches for seismic SSI analysis in nuclear industry. The comparison of ISRS obtained from NLTD models (with single and combined nonlinearity) with those obtained from ELTD indicated a significant effect on response due to energy dissipation through interface sliding and soil nonlinearity, neither of which are captured in typical practice using ELFD approaches. In general, the relative importance of site and interface nonlinearities is a function of contact friction coefficient, soil stiffness, and excitation intensity. Details of this case study further demonstrate that the relative importance of nonlinear effects is siteand case-specific.

This study aims to qualitatively describe the seismic behavior of container cranes. Due to the nature of container crane construction, the portal frame in a crane is highlighted as the controlling aspect for seismic performance. Notably, the effects of the portal frame structural response and the uplift/rocking response are considered in depth. A detailed finite element model is used to consider a case-study container crane to verify and expand our understanding of the seismic behavior. It is found that the coupling between portal frame response and rocking is critical, especially when evaluating performance nearing ultimate. There is a narrow margin between first yielding and potential structural collapse. Several general strategies are proposed to design safer, more resilient cranes in the future.

Seaport activities play an important role in local and national economic growth, but are often located in areas of significant seismic activity. Because container wharf systems are responsible for the critical task of loading and unloading cargos, it is important to understand the factors which influence their seismic response. This paper studies the specific effect of wharf-crane interaction for a hypothetical wharf structure as part of the NEES-GC Seismic Risk Mitigation of Ports. Previous studies concluded that during earthquake events, container cranes may act as tuned mass dampers for the underlying wharf structures and thus suggest that it may be appropriate to ignore the dynamic interaction so long as some small portion of the crane's seismic mass is included in the wharf model. However, this conclusion was made based on a simplified wharf model. This study, on the other hand, utilizes a detailed model of a hypothetical wharf including nonlinearities in the piles, pile-deck connections, and soil structure-interaction. Subjecting the wharf and crane to timehistories of ground deformation and pore water pressure, it is demonstrated that considering the crane not only does not reduce wharf response in all cases but may actually amplify structural wharf response. These results suggest that crane-wharf interaction should be given more careful thought than is currently required when evaluating the seismic response of a wharf system.