Role of SSI on seismic performance of nuclear reactors: A case study for a UK nuclear site

https://doi.org/10.1016/j.nucengdes.2020.110691Get rights and content

Highlights

  • Probabilistic seismic risk assessment of nuclear reactor in a low-seismicity region.

  • Effect of soil compliance on seismic performance of nuclear reactor.

  • Effect of soil compliance on probability of unacceptable performance estimates.

Abstract

The role of soil-structure interaction (SSI) on the risk of unacceptable performance of five non-structural components of a nuclear reactor is investigated through a representative case study consisting of a pressurized-water reactor sited at Holyhead, United Kingdom. The seismic hazard at the site is defined by a 10,000 years uniform hazard spectrum, and its corresponding design response spectrum computed for a target structural performance goal of 1 × 10−5. Nine hypothetical soil profiles, with shear wave velocities ranging from 300 m/s to 1,200 m/s, are considered. Non-linear site response analyses are performed to determine the input motion at the foundation level; seismic capacity and demand of the components are computed using response-history analyses. To obtain high-confidence performance estimates from a limited number of results, Monte-Carlo simulation is used to expand the seismic demand matrix. It is shown that the probability of unacceptable performance of the reactor is highly sensitive to SSI effects. Specifically, the beneficial effects induced by SSI on the seismic performance of the reactor are limited to models sited on stiffer soil deposits (ie, Vs ≥ 900 m/s), whereas, the median probability of unacceptable performance increases for medium-to-soft deposits.

Introduction

Studies on soil-structure interaction (SSI) have been the subject of extensive research since the 1970s (Bielak, 1974, Veletsos and Meek, 1974, Johnson, 1980, Ciampoli and Pinto, 1995, Elnashai and McClure, 1996, Mylonakis et al., 1997, Mylonakis and Gazetas, 2000, Lou et al., 2011). The prevalent view considers SSI effects beneficial unless there are layers of sharply contrasting stiffness or soft deposits. For high-level critical infrastructure (e.g., nuclear power plants), such a simplistic assumption cannot be made and SSI effects need to be explicitly considered and quantified (Kammerer et al., 2018, Pecker et al., 2019). Howard (1973) showed that neglecting the non-linear effects introduced by SSI may lead to non-conservative estimates. Scavuzzo and Raftopoulos (1974) concluded that SSI effects alter the seismic response of both structural and non-structural elements of a nuclear reactor building. Ganev et al. (1997) introduced a simplified sway-rocking model to account for the soil-structure interaction of a nuclear reactor sited on a stiff soil in Hualien, Taiwan, and showed good agreement between numerical estimates and real measurements. Nakagawa et al. (1998) used a modal-equivalent stick model with compliant soil to simulate results obtained from forced vibration tests; it was shown that the proposed model successfully captured the main dynamic characteristics of the reactor building. Nakamura et al. (2010) compared fragility curves obtained from 3-dimensional finite element model and equivalent lumped mass model, concluding that latter tended to overestimate the damage state. Jeremić et al. (2013) highlighted the importance of effects due to soil/rock layering and foundation embedment. More recent studies have focused on the issue of defining the seismic demand in regions of low-to-moderate seismicity, where the lack of real strong motions and tectonic uncertainties pose additional challenges to the design and assessment of nuclear facilities which, differently from ordinary buildings, need to be designed against stringent seismic design criteria considering low-probability of occurrence earthquakes, with return periods as high as 10,000 years. Medel-Vera and Ji, 2016a, Medel-Vera and Ji, 2016b developed stochastic ground motion accelerograms to be used for seismic risk assessment of nuclear reactors in regions characterised by a paucity of real strong ground motions. Yet the use of artificial accelerograms is only recommended for elastic response-history analyses, whereas actual recorded earthquake ground motions, or modified recorded ground motions, are desirable for nonlinear seismic analyses (ASCE/SEI 43-05, 2005). Yawson and Lombardi (2018) investigated the seismic performance of secondary components of a nuclear reactor sited in a hypothetical site in the UK under different seismic hazard definitions. It was found that artificial accelerograms led to non-conservative results for the considered case.

The present work investigates the effects of soil-structure interaction on the seismic performance of key components of pressurized-water reactor sited at Holyhead in the United Kingdom (UK). The term reactor is hereafter used to refer to the containment structure and its internal structure. The seismic assessment adopted in this study follows the methodology set out by Huang et al., 2008, Huang et al., 2011a, Huang et al., 2011b, which builds upon traditional seismic risk assessment methods (Kennedy et al., 1980, Smith et al., 1981, Kennedy and Ravindra, 1984, RA-S, ASME, 2008). The engineering demand parameter is evaluated based on floor spectral accelerations recorded at the different nodes representing the key components. To obtain high confidence seismic performance estimates from a limited number, the demand matrix obtained from inelastic response history analyses is expanded through a Monte-Carlo Simulation. Effects of SSI on the components’ performance are finally quantified in terms of cumulative probability of unacceptable performance distributions and their corresponding median values.

Section snippets

Design response spectra and selection of real accelerograms

The study site is located at Holyhead, United Kingdom. The chosen location is home to two 490 MW Magnox nuclear reactors, and there are plans to build a new £13bn advanced boiling water reactor. A probabilistic seismic hazard assessment for Holyhead by Goda et al. (2013) provides the 5%-damped uniform hazard spectrum (UHS) at bedrock. Fig. 1 shows the UHS for 1,000 and 10,000 return periods that were used to determine the design response spectrum (DRS) for the site. Following the provisions of

Nuclear reactor model

The containment and internal structures of the reactor were modelled using a lumped-mass stick model as shown in Fig. 6. Five key secondary components, namely reactor assembly, feeders, heat transport system, steam generator, maintenance crane, were modelled by means of lumped masses attached at different elevations of the internal structure. The numerical model was developed in the OpenSees platform (Mazzoni et al., 2006) using elastic Timoshenko beam elements and the modified

Conclusion

Following the 2011 Fukushima Daiichi nuclear disaster, there has been an increasing attention to the risk posed by low-probability of occurrence earthquakes on the seismic performance of nuclear facilities, including in regions characterised by low-to-moderate seismicity. This has momentarily halted the construction of new nuclear power plants, and even phased-out nuclear power generation in some countries like Germany. Governments, stakeholders and regulatory bodies have called for greater

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

Authors wish to thank Professor Katsu Goda from the University of Western Ontario, Canada, for providing the data to construct the Uniform Hazard Spectrum shown in Fig. 1, and results from the deaggregation analysis shown in Fig. 2.

References (48)

  • S. Nakagawa et al.

    Forced vibration tests and simulation analyses of a nuclear reactor building

    Nucl. Eng. Des.

    (1998)
  • N. Nakamura et al.

    Study of ultimate seismic response and fragility evaluation of nuclear power building using nonlinear three-dimensional finite element model

    Nucl. Eng. Des.

    (2010)
  • R.J. Scavuzzo et al.

    A review of soil-structure interaction effects in the seimic analysis of nuclear power plants

    Nucl. Eng. Des.

    (1974)
  • P.G. Somerville

    Magnitude scaling of the near fault rupture directivity pulse

    Phys. Earth Planet. Inter.

    (2003)
  • P.Y. Yawson et al.

    Probabilistic seismic risk assessment of nuclear reactor in a hypothetical UK site

    Soil Dyn. Earthquake Eng.

    (2018)
  • J.A. Abell et al.

    Earthquake soil-structure interaction of nuclear power plants, differences in response to 3-D, 3× 1-D, and 1-D excitations

    Earthquake Eng. Struct. Dyn.

    (2018)
  • S. Akkar et al.

    Reference database for seismic ground-motion in Europe

    Bull. Earthq. Eng.

    (2014)
  • ASCE/SEI 43-05

    Seismic Design Criteria for Structures, Systems, and Components in Nuclear Facilities

    (2005)
  • J. Bielak

    Dynamic behaviour of structures with embedded foundations

    Earthquake Eng. Struct. Dyn.

    (1974)
  • M. Ciampoli et al.

    Effects of soil-structure interaction on inelastic seismic response of bridge piers

    J. Struct. Eng.

    (1995)
  • R. Courant et al.

    On the partial difference equations of mathematical physics

    IBM J. Res. Dev.

    (1967)
  • R. Dobry et al.

    Simplified procedures for estimating the fundamental period of a soil profile

    Bull. Seismol. Soc. Am.

    (1976)
  • A.S. Elnashai et al.

    Effect of modelling assumptions and input motion characteristics on seismic design parameters of RC bridge piers

    Earthquake Eng. Struct. Dyn.

    (1996)
  • Gajan, S., Hutchinson, T.C., Kutter, B.L., Raychowdhury, P., Ugalde, J.A. and Stewart, J.P., 2008. Numerical models for...
  • Cited by (4)

    • Probabilistic seismic performance evaluation of nuclear containment structure subjected to far-fault ground motions

      2021, Structures
      Citation Excerpt :

      Hoseyni et al. [22] conducted seismic fragility analysis for a typical containment structure and soil structure interaction is considered. Mistry et al. [23] studied the effect of soil-structure interaction (SSI) on the risk of unacceptable performance of five non-structural components in nuclear power plant. Ma et al. [24] studied the influence of frequency content of ground motions on the seismic fragility of nuclear power plant (NPP) equipment using the transfer functions method.

    • On the seismic analysis and design of offshore wind turbines

      2021, Soil Dynamics and Earthquake Engineering
      Citation Excerpt :

      Several studies are available in the literature having different trade-off between modelling detailing and computational efforts. Some studies investigated the structural components (e.g. Refs. [15,17]) and only a few investigated the geotechnical interaction thoroughly under seismic actions [18]. Soil-structure interactions of piles is best modelled using Beam on Non-linear Winkler Foundations (BNWF) model [10] where the interaction is modelled by a set of springs defined by so-called p-y curves, where p denotes the pressure exterted by the soil and y is the lateral deflection of the pile.

    View full text