Abstract
Seismic protection systems (SPS) have been developed and used successfully in conventional structures, but their applications in nuclear power plants (NPPs) are scarce. However, valuable research has been conducted worldwide to include SPS in nuclear engineering design. This study aims to provide a state-of-the-art review of SPS in nuclear engineering and to answer four significant research questions: (1) why are SPS not adopted in the nuclear industry and what issues have prevented their deployment? (2) what types of SPS are being considered in nuclear engineering research? (3) what are the strategies for location of SPS within NPPs? and (4) how may SPS provide improved structural performance and safety of NPPs under seismic actions? This review is conducted following the procedures of systematic reviews, where possible.
The issues concerning the use of SPS in NPPs are identified: cost, safety, licensing and scarcity of applications. NPPs demand full structural integrity and reactor's safe shutdown during earthquake actions. Therefore, horizontal isolation may be insufficient in active seismic zones and isolation in the vertical direction may be required. Based on the results in this review, it is likely that next generation reactors in seismic zones will include state-of-the-art SPS to achieve full standardised design.
1. Introduction
Seismic protection systems (SPS) are currently considered to be part of a proven and mature technology for mitigating the effects of seismic actions in a wide range of civil structures [Citation1,2]. However, there are currently only two nuclear power plants (NPPs) equipped with such technology, and these were designed about 40 years ago [Citation3]. There is a clear difference between developments in two areas: on the one hand, the development of SPS and their successful application in controlling seismic effects in a broad variety of civil structures; on the other hand, the growing nuclear engineering industry, whose evolution has not included SPS to provide increased seismic safety in its structures. Despite the fact that no NPP has been built equipped with SPS in the last 40 years, extensive research has been conducted in different countries in order to include seismic protection technologies in nuclear engineering design. The first review of seismic isolation for nuclear structures was published 35 years ago [Citation4]. The last peered literature review on this topic was published over 20 years ago [Citation5]. This work aims to provide an update review and includes publications from 1980s to 2010s and answer four specific review questions.
This review is conducted following the procedure of a systematic review [Citation6], where possible. This methodology is often used in health care and medical research [Citation7,8], and is normally used to assess a wider range of experiments, interventions or trials. A systematic review can be seen as a process, in which a group of protocols, checklists or procedures are followed during a literature review in order to answer specific questions. As it is systematic, a similar outcome can be reached by other researchers following the same or similar procedures. At the end of the process, a sound body of evidence has been assembled based on the best information available. This should allow the original questions to be answered and provide a foundation for a critical analysis of the existing research.
All of the general procedures of a systematic review are followed, except the realisation of meta-analysis, as it is not applicable in this qualitative research. The modified methodology consists of five steps.
1.1. Step 1: review questions
Establishing clear questions is a fundamental requirement for defining the framework of the review. In this light, four key questions have been identified and used in the present work: (1) why are SPS not used in the nuclear industry and what issues have prevented their deployment in NPPs? (2) what sort of seismic protection devices are being considered in nuclear engineering research? (3) what are the strategies for the location of SPS within NPPs? and (4) how may SPS help to improve structural performance and safety of NPPs under seismic actions? This work aims to answer the four questions and identify areas for further research through providing a literature review of SPS in nuclear engineering research.
1.2. Step 2: search strategy
A search strategy needs to be defined, comprising keywords/descriptors and sources of information. For this work, the keywords used for the initial search are listed in .
Table 1. Key words and descriptors used to define the search strategy.
The sources of information used for this work are the international peer-reviewed journals listed in and the recent conference proceedings on nuclear/earthquake engineering shown in . The proceedings of the latest conferences on Structural Mechanics in Reactor Technology (SMiRT) held in 2011 (New Delhi, India) and 2013 (San Francisco, CA, USA) were not available for public domain at the moment of submission of this work. The search provides more than 350 papers. In addition, four books about core subjects for this work (both nuclear energy and seismic protection technologies) were consulted and three books about systematic reviews were also included.
Table 2. Journals examined in the search.
Table 3. Conference proceedings examined in the search
1.3. Step 3: selection criteria
The abstracts and conclusions of the selected papers were carefully read and their body scanned and read to different degrees. Selection criteria were set to discard those studies which were not directly related to any of the four review questions. The outcome of this step resulted in 80 relevant papers. With the seven books this gave a total of 87 references selected for this work.
1.4. Step 4: individual findings
Each reference was analysed, its contribution was extracted and presented in standardised tables. In this way, a body of information was constructed aiming to ease the interpretation of results in the next stage. and are the outcome of this step.
Table 4. Example of devices for different protection approaches.
Table 5. Key differences in seismic design of SPS: NPPs vs. conventional structures.
Table 6. Summary of research on SPS for nuclear islands.
Table 6. (Continue).
Table 6. (Continue).
Table 6. (Continue).
Table 7. Summary of general mechanical and geometric features of devices.
1.5. Step 5: analysis and interpretation
Based on the reviewers’ judgement, all the evidence/information during the process are gathered and analysed in such a way that the review questions can be appropriately answered. This step forms the main body of this review and is presented in five sections:
Section 2 is aimed at identifying links between the areas of seismic protection technologies and the nuclear industry and examining how these aspects affect the low acceptance of SPS in NPPs. This section attempts to answer the first review question.
Section 3 aims to characterise the devices in SPS for use in nuclear facilities. This characterisation is made in terms of physical (e.g. geometry, material, etc.) and mechanical (e.g. stiffness, damping, etc.) properties, identifying similarities and differences in comparison with those considered in conventional structures. This section contributes to answering the second review question.
The purpose of Section 4 is to establish the spatial configuration (e.g. two-dimensional (2D), three-dimensional (3D), plan and height distribution, etc.) of how the seismic protection is considered for nuclear facilities, recognising similarities and differences in comparison to approaches used in traditional civil structures. This section is intended to answer the third review question.
As most of the research conducted relates to the safety-related parts of a nuclear facility, Section 5 aims to complement the answers presented in the last two sections by addressing SPS for non-safety-related equipment of NPPs.
Section 6 has two objectives. First, the establishment of how the devices (Section 3) and spatial configurations (Section 4) may provide a substantial increment in seismic safety for NPPs. This part attempts to answer the fourth review question. Second, this section covers future research and identifies new research questions on this topic.
2. Key issues for seismic protection systems of nuclear power plants
2.1. Development of SPS for NPPs
The deployment of SPS, and particularly seismic isolation in NPPs, started in the 1970s with two facilities: Koeberg NPP in South Africa, which had two isolated unit plants; and Cruas NPP in France, which had four isolated unit plants [Citation3,Citation9]. After this promising start, various countries with a wide range of seismic activities reported research on the applicability of SPS in future nuclear power stations. During the 1980s in Japan, a joint commercial industry project began long-term research on the feasibility of the use of seismic isolation in combination with energy dissipation devices for light water reactors (LWRs) [Citation10,11]. This is currently the most common nuclear reactor in the world, comprising 80% of the units worldwide [Citation12]. Also during the 1980s, a joint government, industry and academia project in the USA started to develop a comprehensive Research and development (R&D) programme on the applicability of seismic isolation into NPPs [Citation13]. By the end of 1980s, research efforts were focused on developing the first seismically protected American nuclear reactor, the Advanced Liquid Metal Reactor (ALMR), supported by high-damping rubber bearings (HDRB). This project aimed to develop a standardised reactor building in order to make it seismically safe and economically competitive [Citation14]. At the same time, similar efforts were made in Italy for the development of innovative nuclear reactors, called RSP/I by their Italian initials, which would provide high levels of seismic safety by the addition of HDRB. Again, the main objective of this initiative was the standardisation of their design in order to reach not only rigorous safety levels, but also economic competitiveness [Citation15]. Even countries with low to medium seismicity reported investigations with the same objectives. In the early 1990s, the United Kingdom was part of a tripartite agreement of power industries, alongside the USA and Japan, contributing through experimental research on the behaviour of laminated rubber bearings and viscous dampers applicable to liquid-metal-cooled reactors (LMR) [Citation16]. Similarly, by the mid-1990s, a long-term R&D project was developed in South Korea intending to determine the effectiveness of different types of laminated rubber bearings for the nuclear industry, focused on their applicability to the Korea Advanced Liquid Metal Reactor (KALIMER) [Citation17]. During mid-1980s, the local experience in New Zealand on seismic isolation was reported based on lead–rubber bearings (LRBs) and their potential application to nuclear reactors [Citation18].
A common feature of the earlier work and many subsequent research projects was the standardised seismic design of the so-called ‘nuclear island’. This is formed by the containment vessel and internal safety-related components, including the reactor vessel, all of them supported by a common foundation structure [Citation19,20]. A standardised seismic design using SPS is aimed at the possibility that a NPP would become site-independent of the local seismicity of the construction location. This may promote the deployment of the same NPP design in any geographical region, regardless of how severe their seismicity levels are. The potential benefits of a standardised seismic design provided by the use of SPS have a direct impact on the following issues: (1) reducing costs, both in the amount of structure needed and the seismic equipment qualification, be it analytical and/or experimental proof of the adequacy of safety-related equipment to withstand seismic loads; (2) a substantial increment in safety margins including the assurance of full structural integrity after a severe earthquake; and (3) an improvement of economic competitiveness in comparison with other sources of energy and simplification of licensing procedures [Citation5,Citation15,Citation16,Citation19,Citation21–23]. However, it seems that the limited number of real applications of seismically protected NPPs and the lack of specific codes and standards to design SPS for nuclear structures are the key reasons in preventing their deployment [Citation3,Citation23].
The nuclear industry is developing safer and more efficient nuclear reactors, but the systems become increasingly sensitive to seismic loads. Coladant [Citation9] and Gantenbein and Buland [Citation24] pointed out that fast breeder reactors (FBRs), belonging to the so-called Generation III/III+, were more sensitive to seismic loads than pressurised water reactors (PWRs) belonging to the Generation II. Lo Frano and Forasassi [Citation25,26] pointed out that the primary coolant in a lead-cooled fast reactor (LFR, in the category of Generation IV) is a high-mass-density fluid and additional inertia forces (sloshing) should be taken into account to ensure an adequate seismic response of the reactor. In addition, Austin et al. [Citation16], Kato et al. [Citation27] and Okamura et al. [Citation28] pointed out that modern reactors were increasingly subjected to larger thermal loads in comparison with previous versions. One way to reduce this thermal stress intensity could be achieved by using thinner structural elements, which in turn may lead to an unsatisfactory performance of the nuclear island under seismic loads. In this light, SPS provide a solution, compromising between the thinner elements and better seismic mitigation features, in comparison with the design of previous generation rectors. It is likely that the next generation of reactors in development, Generations III and IV, will be more demanding in terms of their serviceability requirements under seismic actions. Consequently, SPS may play a major role in assuring this functionality requirement in NPPs.
The following subsections describe the four issues identified by the seismic protection field and the nuclear industry and how they relate to each other.
2.2. Cost
Due to the limited real applications of seismic isolated NPPs, it is difficult to provide sound remarks about cost. Nevertheless, some insightful considerations have been reported in the literature. It was early suggested that, in terms of total project costs, no major differences can be expected between a seismically isolated NPP and the non-protected version of the same facility [Citation29–31]. More recently, Malushte and Whittaker [Citation23] reported concerns about the exceptionally high construction costs of next generation NPPs, and the use of SPS would reduce the overall schedule due to a standardised seismic design. For a seismically protected NPP, it is necessary to consider the costs of the SPS plus the elements required to fit the external devices and the seismic gap elements required to accommodate displacements with adjacent non-isolated structures. However, there is also the reduction of costs related to the lower seismic design loads, which would lead to a reduction in the size of the main structural elements and the seismic qualification of safety-related equipment [Citation18,19]. It is acknowledged that seismic loads may not control the design of NPPs, which is more likely to be controlled by radiation shielding and/or loss of coolant accident. However, this does not necessarily apply for many safety-related equipment [Citation23].
The cost of a generic SPS depends on the level of protection required, and therefore, its complexity. However, it is necessary to consider the addition of certain sub-structures, for example, an upper raft acting as a support for the entire nuclear island. This is required in order to decouple the nuclear island from the ground, considering the presence of the SPS between the upper raft and the lower raft (foundation). Some results for non-nuclear applications reported that a seismic isolation system may be around 2%–4% of the overall cost of the building [Citation29]. Similar figures were reported by Staudacher [Citation31] who pointed out a cost of about 5%–10% of raw construction, which in turn is roughly one-third of the total building cost. As the equipment inside a nuclear facility is expected to account for most of the total cost of the facility, lower relative percentages are to be expected than for non-nuclear buildings. Skinner et al. [Citation32] estimated that the cost of generic SPS was less than 2% of the overall cost of an NPP.
Reduction of construction costs is directly related to the consumption of concrete and steel reinforcement. Kato et al. [Citation10] reported savings between 5% and 10% in a seismically isolated boiling water reactor (BWR) building in comparison with the non-isolated version. For a non-nuclear application, Buckle [Citation18] reported similar figures, reaching savings of 10% in material costs. It is acknowledged that the discussion above is based on the promising early steps of nuclear engineering research. No updated information about costs has been found in the literature for this review. However, a related scenario is on non-nuclear applications. De la Llera et al. [Citation33] reported that two seismically isolated high-critical hospitals had costs in the same range of conventional designs.
Another aspect that should not be overlooked is related to the engineering fees. Eidinger and Kelly [Citation21] stated that it was possible to reach savings between 40% and 60% for this item in the design of a standardised NPP in contrast with a non-standardised nuclear facility, considering equal power capacity. These savings were related not only to the significant simplification of the design of the primary earthquake-resistant elements, but also, perhaps even more importantly, to the design of connections/fixings of safety systems, equipment, pipes and components of the nuclear island [Citation23].
A further aspect regarding costs issues, which seems not to be covered in literature, is related to the costs of re-launching the NPP after the occurrence of a relevant earthquake. It will be seen in Section 2.2, using SPS is the most reliable way of assuring full structural integrity and reactor's safe shutdown after a seismic action of medium to large intensity. As traditional designs may not be able to provide the structural reliability required for a nuclear facility, both structural and non-structural damages may be expected under seismic actions. Of course, the latter is related to a certain economic loss to the nuclear facility itself, but the concern is the possible occurrence of a nuclear accident. At this point, further aspects should be taken into account, such as the impact of stopping the NPP as an energy supplier and, certainly, the potential danger towards the environment. It is important to note that in realistic cost analysis, not only the direct cost of the SPS itself should be considered, but also the difference in the structural reliability provided by a seismically protected nuclear facility that is able to ensure full structural integrity after medium and severe earthquakes.
2.3. Safety
NPPs and energy supply facilities require higher levels of safety than conventionally designed buildings as they play a strategic role after the occurrence of a severe seismic event [Citation31]. In this light, seismic protection technologies offer a feasible method of reaching a higher level of protection by adding special devices within the structure. These technologies are designed to take most of the inelastic deformations imposed by earthquakes, allowing the superstructure to remain essentially in its elastic range [Citation18, Citation34-35] as its seismic demand could be reduced 6–8 times when compared to a conventional structure [Citation17]. Therefore, SPS improve the overall safety and reliability of the building, contributing to ensuring full structural integrity and operability after relevant seismic events [Citation36]. It is important to realise that for conventional seismic design the situation is different. In such a case, structures dissipate energy through inelastic deformation located at plastic hinge zones, and therefore, structural and non-structural damages are allowed to a certain extent as long as the condition of life safety is ensured. However, for NPPs this would lead to unacceptable seismic performance.
It should be noted that the addition of SPS to confine the inelastic deformation of a structure has a beneficial spin-off in terms of engineering modelling and design. For a conventional structure, the study of its inelastic dynamic response is based on assumptions about the non-linear behaviour of its structural elements. Such a situation becomes more complex for NPPs as additional non-linear constraints may affect critical components due to severe thermal loads [Citation27,Citation32]. On the other hand, in a seismically isolated building, the location where the non-linear behaviour will take place is known – the SPS – and therefore, the inelastic response of the structure will also be known. In fact, this may be explained due to the sum of two effects: (1) the possibility of having devices fully tested in the laboratory and hence, knowing their hysteresis curves; and/or (2) the modification of the dynamics of the structure imposed by the SPS which moves the structure to a condition where seismic loads are smaller, and therefore, ensuring the response of the superstructure is in its elastic range [Citation18,Citation34,Citation35]. This fact was early recognised for the particular example of horizontal isolation provided by laminated rubber bearings (e.g. [Citation19,Citation31]). These devices provided an interface of very low stiffness between the superstructure and the foundation, whilst being able to withstand most of the displacement imposed by earthquakes. Consequently, the superstructure is subjected to a reduced seismic input and behaves essentially as a rigid body [Citation26,Citation37]. This is also advantageous for making simpler structural models in preliminary or feasibility analyses.
In general terms, the seismic behaviour of a building with SPS can be better predicted by structural models than that without SPS [Citation5]. It must be noted that uncertainties of the seismic input are still present in both cases. Nevertheless, this aspect should not be overlooked as it will benefit the levels of accuracy of structural modelling and seismic design, and consequently, the overall safety and reliability of NPPs under seismic loads [Citation22].
2.4. Applications
The particularly small number of real nuclear applications may have prevented the general acceptance of SPS in the development of nuclear structures. Even in the early stages of seismic isolation technology, the concept itself was questioned. Hadjian and Tseng [Citation38] pointed out that the few field experiences of isolated applications subjected to design conditions were not enough to validate its use. Hence, the transference of the applicability of seismic isolation into nuclear industry would need further and stronger evidence. Nevertheless, in more recent developments of seismic protection, Forni et al. [Citation3] and Martelli [Citation2] reported a different scenario: a general consensus was reached among the technical community, in which seismic protection techniques had become a mature and reliable technology in mitigating seismic effects over a wide variety of civil structures. They pointed out the existence of about 10,000 applications of seismically protected structures in countries such as Japan, USA, Italy, China, Russia and Chile, ranging from areas of high seismic activity to countries of moderate and low seismicity. Some of these applications have already been subjected to severe seismic motions, achieving acceptable structural performances and providing the condition of full structural integrity. Furthermore, it is important to highlight that a key contribution to real nuclear applications equipped with SPS will be made in the near future. Two reactors, although for experimental and research purposes, are currently under construction equipped with seismic isolators in Cadarache, France, an area with moderate seismic activity. The Jules Horowitz Reactor (JHR), aimed at research in nuclear medicine, and the International Thermonuclear Experimental Reactor (ITER), intended for experimental activities in fusion energy, are the new real nuclear applications seismically protected after Koeberg NPP and Cruas NPP [Citation3].
Although the number of NPPs seismically protected is still small, significant research efforts have been recently made in order to broaden the number of applications of NPPs equipped with SPS. There are two significant examples: (1) Takahashi et al. [Citation39] reported that the Japanese government sponsored a large scale R&D project on 3D seismic isolation systems for FBRs between 2000 and 2005, in an attempt to enhance and generalise the concept of seismic protection; (2) Forni and De Grandis [Citation40] summarised comprehensive research conducted in the framework of the SILER Project (Seismic-Initiated Events Risk Mitigation in Lead-Cooled Reactors) that is aimed at implementing seismic isolation on Generation IV heavy metal reactors. These initiatives, among others, will also provide the opportunity to deploy nuclear power in more areas, such as developing countries and other nations with increasingly higher demands for energy.
2.5. Licensing
Major licensing hurdles have made it difficult to fully deploy SPS in the nuclear industry. Such hurdles may be explained mainly due to the lack of specific codes and standards on seismically protected NPPs [Citation3]. Additionally, it seems there is a general tendency for nuclear industry practitioners to keep the traditional approach in the seismic design of NPPs and/or an apparent lack of sound knowledge amongst the practitioners concerning the latest seismic protection technologies [Citation23]. These facts could have discouraged the realisation of projects of seismically protected nuclear stations, and consequently, led owners to take traditional design approaches in order to not jeopardise entire projects.
There are several encouraging initiatives towards the development of codes and standards for design of NPPs equipped with SPS. The following examples can be cited: (1) in Japan, the code JEAG 4614-2000 Technical Guideline on Seismic Base Isolated System for Structural Safety and Design of Nuclear Power Plants is the only standard that specifically addressed the seismic isolation of NPPs [Citation3]; (2) in the USA, the standard ASCE 4-98 Seismic Analysis of Safety-Related Nuclear Structures includes a section addressing requirements for design of isolated nuclear structures, although it has not yet endorsed by the American nuclear regulatory authority [Citation23]; (3) in Europe, the report EUR 16559 Proposal for Design Guidelines for Seismically Isolated Nuclear Plants [Citation41]; and (4) the International Atomic Energy Agency's report IAEA-TECDOC-1288 Verification of Analysis Methods for Predicting the Behaviour of Seismically Isolated Nuclear Structures [Citation34].
Recent seismic probabilistic risk analyses have demonstrated that the use of SPS drastically reduces the seismic risk of NPPs [Citation42]. This fact should encourage the reduction and simplification of licensing procedures. Certainly, the initial acceptance of seismic protection itself is a major issue for the established nuclear industry [Citation30]. However, this licensing hurdle may be considered only as an initial barrier, as once the first licence is awarded, it is expected that the process of obtaining the following ones will be easier and faster [Citation21]. This will also increase the economic competitiveness of nuclear energy generation compared with other sources of energy [Citation15]. Considering the fact that SPS have made important progress in technical and non-technical aspects in recent years, it seems that the nuclear engineering industry should be dedicated to address all industry/regulatory issues pending in order to reach the first new generation of NPPs equipped with state-of-the-art SPS [Citation23].
3. Devices in seismic protection systems for nuclear power plants
3.1. Development of devices in SPS for NPP
Seismic protection can be provided through three approaches: (1) passive, (2) semi-active and (3) active. Passive systems are the devices located within a structure aimed to increase the inherent energy dissipation capacity of the structure. These systems do not have the ability to change their dynamic properties during seismic excitations and they do not generate active control forces to the structure. A semi-active system is able to change its dynamic properties during a seismic excitation based on feedback provided by the monitored response of the structure, but they do not generate control forces to the structural system. Active systems have the ability to adjust their dynamic properties, provided by the feedback of the real-time response of the structure, and to apply active control forces to the structure, through mechanical or hydraulic actuators. These systems need a well-defined control strategy and use the dynamic response of the structure to determine appropriate control signals to be sent to the actuators [Citation43]. For nuclear engineering, it is noticeable that the majority of research, both theoretical and experimental, is aimed at the use of passive systems, because it is more practical than the active or semi-active control mechanisms. A few cases of active systems have been reported only at a preliminary level whereas little attention has been paid to semi-active systems. gives examples of devices for each approach [Citation44].
The devices listed in are suitable for deployment over all ranges of seismicity levels for either a non-critical building or a high-risk civil structure. Therefore, their governing principles and design process remain essentially unchanged. Nevertheless, NPPs have distinctive requirements that make them different from conventional structures. Consequently, the goals of seismic protection between both types of structures are different [Citation45]. The key differences are indicated in .
It is inferred from that there are two issues related to using SPS in nuclear facilities to match their required seismic performance. (1) Redundancy: SPS for a conventional building normally rely on one group of traditional devices (e.g. elastomeric bearings); for NPPs, it may be necessary to use two or more groups of traditional devices (e.g. elastomeric bearings plus viscous fluid dampers) and/or the inclusion of a fail-safe system against a beyond-design basis event. (2) Spatial configuration: SPS for non-critical buildings do not normally incorporate isolation in the vertical direction; new devices able to deal with the vertical vibration may need to be considered, as well as their location within the structure. The following subsections provide brief details of the devices for seismic protection suitable for nuclear engineering. summarises the research on SPS to protect the entire nuclear island. In this table, column 1 gives the source of information; column 2 provides the type of devices considered and their spatial configuration, aiming to provide either horizontal isolation (2D) or full isolation (3D) (to be discussed in the next section); column 3 indicates the application of NPP considered and their potential country of deployment (if applicable) and column 4 highlights the main achievements.
3.2. Passive devices
3.2.1. Elastomeric bearings
Elastomeric-based bearings, also known generically as seismic isolators or laminated rubber bearings, are composed of layers of either natural rubber or neoprene with alternated steel plates bonded by vulcanisation. These devices have been widely used in seismic protection of non-critical buildings, and comprehensive research has been conducted on their application in the nuclear industry. Depending on the kind of elastomer used, they can be identified as low-damping rubber bearings (LDRB) or HDRB [Citation66]. (a) shows a schematic view of an elastomeric bearing and (b) gives a typical hysteretic curve of an HDRB device. General features reported for these devices for potential nuclear applications can be seen in .
Figure 1. Sketch and mechanical behaviour of elastomeric bearings [Citation62]: (a) schematic view; (b) hysteretic behaviour.
![Figure 1. Sketch and mechanical behaviour of elastomeric bearings [Citation62]: (a) schematic view; (b) hysteretic behaviour.](/cms/asset/7c1b03d9-fa79-400a-ba8f-a30490496cb2/tnst_a_980347_f0001_oc.jpg)
3.2.2. Lead–rubber bearings
LRBs are also elastomeric-based devices, normally made using low-damping natural rubber. The difference between basic elastomeric bearings and LRBs is the addition of a lead plug or cylinder which in turn enhances the damping capacity of the bearing [Citation18,Citation62,Citation66]. These devices, initially devised in New Zealand, have also been widely used in the seismic protection of conventional buildings. Their application to nuclear facilities has also been considered by researchers. (a) shows a schematic view of an LRB device and (b) gives a typical hysteretic loop. The latter can be well represented by a bilinear model [Citation18] providing the mechanical behaviour of the lead plug (elasto-plastic) and the lateral response of the natural rubber (linear). General features reported for these devices for potential nuclear applications can be seen in .
Figure 2. Sketch and mechanical behaviour of lead–rubber bearings [Citation18,Citation62]: (a) schematic view; (b) hysteretic behaviour.
![Figure 2. Sketch and mechanical behaviour of lead–rubber bearings [Citation18,Citation62]: (a) schematic view; (b) hysteretic behaviour.](/cms/asset/33926363-52d7-4063-a7c4-aa5b6e4afb6a/tnst_a_980347_f0002_oc.jpg)
It is worth mentioning that more than 75% of the articles referenced in considered the use of elastomeric and LRBs. Therefore, it is highly likely that such devices will be used in new-generation nuclear deployments.
3.2.3. Steel springs
Steel springs are intended to isolate the vertical direction of a structure, and also need (1) to withstand the weight of the structure and to provide a long period of vibration in the vertical direction, (2) to be stiff in the other degrees of freedom and (3) to have a non-brittle failure mode [Citation51]. In line with this, research efforts in nuclear engineering have been placed on three types of devices: coned disk springs, metallic bellows and helical springs. These devices have been considered at experimental level for nuclear deployment.
Coned disk springs are made of high-tensile steel of springs, and single disks can be stacked in different arrangements in series and parallel in order to reach different strokes and energy dissipation levels, representing a versatile alternative for vertical isolation of NPP. Hysteresis is generated by friction between single disks stacked in parallel and between groups of disks stacked in series and the stiff centre guide, which in turn restrains the other degrees of freedom of the device. The efficient friction of these devices becomes critical in order to reach high levels of energy dissipation required [Citation52]. Their potential deployment in the nuclear engineering industry has been recently investigated in Japan, specifically for their application in FBRs [Citation28,Citation51–53,Citation61]. (a) shows a schematic view of an arrangement of coned disk springs considered for vertical isolation of FBRs and (b) shows the hysteretic behaviour obtained for a particular arrangement of disks considered at experimental level for potential nuclear applications.
Figure 3. Sketch and mechanical behaviour of coned disk springs [Citation61]: (a) schematic view; (b) hysteretic behaviour.
![Figure 3. Sketch and mechanical behaviour of coned disk springs [Citation61]: (a) schematic view; (b) hysteretic behaviour.](/cms/asset/dcbce569-be8f-471c-a76a-d21fcb0c282f/tnst_a_980347_f0003_oc.jpg)
Metallic bellows are composed of multiple thin layers of steel and a certain number of convolutions which determine the vertical displacement capacity of the device. The device comprises one main bellow and one auxiliary bellow subject to high internal pressure. Bellows are wrapped in reinforcement rings in order to provide stability to the whole system, while constraining the other degrees of freedom of the device. Therefore, they can be used as vertical isolation devices as they can work as a low-frequency air spring. They have been considered preliminarily for application for FBRs in Japan [Citation60]. (a) shows a schematic view of an experimental isolator based on metallic bellows to isolate the vertical direction and (b) shows a load vs. displacement relationship in the vertical direction obtained experimentally in a 1/5 scaled model.
Figure 4. Sketch and mechanical behaviour of metallic bellows [Citation60]: (a) schematic view; (b) hysteretic behaviour.
![Figure 4. Sketch and mechanical behaviour of metallic bellows [Citation60]: (a) schematic view; (b) hysteretic behaviour.](/cms/asset/b1c3710b-7d54-44c8-b7ab-0a9f6ae2cbc5/tnst_a_980347_f0004_oc.jpg)
Vertical isolation using steel springs was investigated based on traditional-shaped helical springs. Gantenbein and Buland [Citation24] reported experimental and analytical studies on their potential application in PWRs and FBRs. Incorporating helical steel springs for vertical isolation in critical facilities was also studied in Mexico (NPP) and Germany (chemical plant) [Citation29], although for mitigating the effects of subsidence of soil and not directly for mitigating seismic effects.
3.2.4. Air springs
Air springs have also been considered recently for deployment in nuclear engineering, particularly for FBRs in Japan. Two types of air springs were experimentally studied: (1) vertical air spring [Citation55,56] and (2) 3D air spring [Citation58]. The former was intended to isolate in the vertical direction of the structure whereas the latter comprises horizontal and vertical isolations in one single device, defining a highly innovative device considered as SPS.
Vertical air spring or rolling seal type air spring is composed of an air compartment sealed by rolling rubber which provides the ability to withstand vertical deformations. As air springs have very little or null hysteresis capacity, they need to work together with levelling devices and rocking suppression devices in order to maintain a constant height of the structure at all times [Citation56]. (a) shows a schematic view of an experimental isolator based on vertical air springs to isolate in the vertical direction and (b) shows a hysteresis curve obtained experimentally in a 1/7 scaled model. Little hysteresis is observed in (b) as a result of friction on contact parts indicated in (a).
Figure 5. Sketch and mechanical behaviour of vertical air spring [Citation55,56]: (a) schematic view; (b) hysteretic behaviour.
![Figure 5. Sketch and mechanical behaviour of vertical air spring [Citation55,56]: (a) schematic view; (b) hysteretic behaviour.](/cms/asset/57a853d9-bb9d-4b4c-b78f-d8ef539978d4/tnst_a_980347_f0005_oc.jpg)
Another type of air spring considered for nuclear deployment, specifically for FBR, is the 3D air spring. They provide horizontal and vertical isolation in one single device based on compressed air and are composed of a rubber sheet between two steel cylinders, reinforcing fabric and reinforcing cables. The gap between the inner and outer cylinders provides the horizontal stroke. The gap between the top of the inner cylinder and the upper structure provides the allowable vertical stroke of the device. Theoretically, this air spring does not have any restoring behaviour; it needs to be used together with a rocking suppression system and supplemental damping in the vertical direction [Citation58]. shows a schematic view of the experimental 3D air spring. Although some experimental response curves were reported, no experimental behaviour curve was found for this device.
Figure 6. Sketch of 3D air spring [Citation58].
![Figure 6. Sketch of 3D air spring [Citation58].](/cms/asset/7c9f8d8f-4405-4f16-bd71-21f5a436c092/tnst_a_980347_f0006_oc.jpg)
3.2.5. Viscous dampers
Viscous fluid dampers, having the same principle as ordinary automotive shock absorbers, are devices that dissipate energy through the conversion of mechanical energy into heat as a piston deforms a highly viscous substance, e.g. silicone gel, oil, etc., inside of a damper housing or container [Citation16,Citation44]. These dampers have been widely used as supplemental energy dissipation devices in conventional civil structures [Citation1], and research has been carried out in nuclear engineering in order to consider them in conjunction with other passive devices to meet the higher requirements of NPPs. shows the latest applications of viscous dampers intended for nuclear applications (e.g., [Citation48,49,Citation58]). The main features of viscous dampers are as follows:
Viscous fluid dampers can be classified into two types: (1) open containers and (2) closed containers. The first one, with cylindrical pot fluid dampers and viscous damping walls, dissipates energy by the relative motion of cylinders or plates inside the fluid in an open container. The second one, with orificed fluid dampers and pre-loaded fluid devices, dissipates energy by forcing the fluid to pass through orifices inside of the device. In general, open container dampers have a relatively lower energy dissipation capacity than the closed container dampers. Nevertheless, the latter require the most sophisticated internal design in contrast with the rather simpler open-container-based devices [Citation44]. Experimental results for open container devices showed a high energy dissipation capacity: Austin et al. [Citation16] reported 8%–10% critical damping ratio, whereas Hüffmann [Citation29] showed a range of 20%–30% achievable with this type of device.
For open container viscous dampers for potential nuclear applications, Austin et al. [Citation16] reported that viscous fluids were both frequency and temperature dependent. In general, the damping coefficient diminishes with higher frequencies until it becomes stable over a certain number of cycles of loading. On the other hand, temperature should be analysed under two aspects: (1) fluid local temperature and (2) ambient temperature. When the internal temperature of the fluid increases after few loading cycles, the viscosity decreases and, therefore, a reduction in the energy dissipation capacity is expected. The same negative effect is produced by ambient temperature, as they stated a loss in the energy dissipation capacity of about 50% with an increase in 15 °C in ambient temperature. A controlled working environment is a strong constraint to be considered with open container devices.
For closed container dampers, the hysteretic behaviour depends on the device's internal geometry as well as the fluid's parameters. However, it is certain that a high level of energy dissipation can be achieved with these devices. The internal geometry can be designed and tested in order to optimise their dissipation capacity. Kato et al. [Citation10] and Matsumura et al. [Citation11] reported experimental attempts both in laboratory tests and in non-nuclear real structures, to deploy viscous oil dampers in LWR applications, confirming an appropriate performance for nuclear applications.
3.2.6. Steel hysteretic dampers
Steel hysteretic dampers are devices whose energy dissipation mechanism is based on the inelastic deformation of a metal, normally mild steel or any highly ductile alloy [Citation44]. As well as viscous dampers, steel dampers have been extensively used as SPS in traditional buildings using a wide variety of metallic materials, shapes and sizes [Citation1], all of them using the same principle to dissipate energy. shows the latest applications of steel hysteretic dampers intended for nuclear applications (e.g. [Citation28,Citation51,Citation52]).
Theories of plasticity and viscoplasticity are used to model the behaviour of steel hysteretic dampers. Temperature is not a relevant agent affecting the devices’ behaviour, even though during a severe earthquake an important portion of the energy is dissipated as heat, raising the temperature of the surrounding material. However, that increment of temperature does not affect the mechanical properties of the device. The exception to this behaviour is for lead-based devices as lead is more sensitive to temperature changes. Additionally, fatigue analysis is critical in order to determine the durability of the device [Citation44]. In the nuclear engineering industry, early attempts were reported by Kato et al. [Citation10] and Matsumura et al. [Citation11] to consider hysteretic dampers by means of steel rods or bars as SPS, alongside laminated rubber bearings, for deployment in LWRs.
Sections 3.2.1–3.2.6 provided an overview of the devices reported for potential nuclear applications. Some of their general features are summarised in . Although all values given in were reported in literature, they must be taken as a reference only. Column 1 indicates types of devices; column 2 provides an estimation of the damping level possible to attain, measured in terms of critical damping ratio; column 3 gives estimations of the deformation capacity, measured in terms of shear strain capacity for rubber bearings and strokes for steel and air springs; column 4 indicates rough geometric dimensions; and finally, column 5 provides some general remarks.
3.3. Semi-active and active devices
The use of semi-active devices as SPS in the nuclear industry has not been investigated yet, as neither theoretical nor experimental results have been reported. Nevertheless, research on the use of semi-active devices has been considered in recent years. A number of applications in traditional civil structures have been successfully deployed. A comprehensive review of semi-active control systems for seismic protection of structures was reported by Symans and Constantinou [Citation67]. The use of this type of protection for nuclear deployment should not be discarded, as this technology has increasingly gained acceptance within seismic protection technologies.
Only a few attempts to deploy active protection in the nuclear industry have been considered. On a theoretical basis, Wolf and Madden [Citation64] reported the benefits of using active control systems for protecting the reactor vessel in NPPs. Later, Izumi [Citation68] reported the existence of the first application in Japan of a low-risk building equipped with an active control system, although with a limited capacity to control the dynamic response of the structure under severe seismic loads. By then, the applicability of active control to NPPs seemed restricted as no remarkable research and development had been carried out at that time. More recently, Radeva [Citation43] proposed an active control system for two NPPs in Bulgaria and provided some experimental evidence on its performance. These systems were based on hydraulic actuators distributed within the structure whose active control forces were determined in real time by sensors monitoring the dynamic response of the structure. In simple terms, the system was able to reconfigure its properties in real time in order to ensure the best possible structural protection available within the control strategy defined. The use of these systems, totally or partially in conjunction with passive devices, may have potential for nuclear applications.
4. Spatial configuration of seismic protection systems for nuclear power plants
4.1. General consideration
The devices described in the last section are arranged in different configurations within the structure in order to provide different types and levels of protection for NPPs. In this regard, three different types of spatial configuration of devices have been identified: (1) two-dimensional systems with horizontal isolation provided in one single interface (2D), (2) three-dimensional systems with horizontal and vertical isolation provided in one single interface (3D) and (3) three-dimensional systems with horizontal and vertical isolation provided in two interfaces [2D + V]. Conventional civil structures have been traditionally equipped with 2D systems, whereas the use of 3D systems is not normally considered. For the case of high-risk facilities, no 3D systems have been reported to mitigate the effects of severe earthquakes. Nevertheless, for the case of NPPs, extensive research has been carried out with this objective for reaching a full generalisation of the seismic protection concept. shows schematically the concepts for providing seismic protection to NPPs.
Figure 7. Schematic approaches to provide SPS in NPP [Citation19,Citation27]: (a) one interface (2D or 3D systems); (b) two interfaces (2D + V systems).
![Figure 7. Schematic approaches to provide SPS in NPP [Citation19,Citation27]: (a) one interface (2D or 3D systems); (b) two interfaces (2D + V systems).](/cms/asset/0af7ecd4-16b6-4535-907f-3e4fb11a3107/tnst_a_980347_f0007_oc.jpg)
Isolation in the vertical direction to set up 3D systems has not been widely used mainly due to two hurdles: (1) the SPS, being flexible in the vertical direction, must be able to withstand the total weight of the structure, and (2) the SPS must be able to control rocking modes, which arise as a natural consequence of a fully isolated structure [Citation55,Citation57,Citation63]. For nuclear applications, it is critical to consider the effect of the vertical component of seismic loads as serious problems related to malfunction of internal components and/or structural integrity, such as uplift of fuel assemblies, reactivity change and buckling of the reactor vessel may arise [Citation28,Citation51]. The necessity of incorporating 3D SPS into nuclear facilities was reported by Hadjian and Tseng [Citation38] and Seidensticker [Citation69]. Without addressing the isolation of vertical and rocking modes, seismic design cannot be considered fully standardised, remaining site-dependent. Early attempts of developing 3D SPS for potential nuclear applications were reported by Coladant [Citation9], Gantenbein and Bulant [Citation24], Hüffmann [Citation29] and Tajirian et al. [Citation5]. For a long time, a number of 3D systems for nuclear deployment have been studied, reaching no applicable results [Citation23,Citation59].
As mentioned in Section 3.2.1, traditional seismic isolation based on laminated rubber bearings acts as stiff as a traditional foundation system in the vertical direction, thus, the seismic vertical component is transmitted in its entirety to the structure [Citation51]. Accordingly, seismically isolated structures based solely on elastomeric bearings are almost not affected by rocking inputs/modes, even though a high variation on the vertical load may be expected and must be taken into account during the isolators’ design process [Citation46]. Furthermore, Kageyama et al. [Citation58], Somaki et al. [Citation61] and Yoo et al. [Citation17] pointed out that the vertical response of internal equipment in horizontally isolated structures tended to be amplified in comparison with the non-isolated version of the building. Hadjian and Tseng [Citation38] also suggested that the seismic design of internal equipment in the vertical direction should be the same as that for the horizontal direction.
It is also worth mentioning the selection of an appropriate plan distribution/layout of the devices as SPS. They should be located uniformly under the nuclear island and closer to each other under stress concentration and/or critical zones, e.g. under main walls and/or under the reactor vessel. This is done in order to have a permanent load distribution as uniform as possible acting on the devices [Citation35]. Additionally, it is desirable to minimise the global torsion mode of the nuclear island. In this way, lateral displacements of the nuclear island are likely to be uniform, which would ease the design of adjacent structures and flexible connections running through them. Furthermore, as all SPS need to be periodically inspected, they should be installed with enough room in order to perform any maintenance or replacement work [Citation57].
4.2. 2D systems
Horizontal isolation, or 2D systems, is the most common approach to isolate a structure against earthquake actions and it has been widely used in conventional civil structures. Significant research has been conducted for potential deployment in the nuclear industry. In fact, the only NPP equipped with SPS, although using the outdated technology, is based on these systems. Koeberg NPP (South Africa) and Cruas NPP (France) possessed seismically isolated reactor units based on square-sliding elastomer pads and squared elastomer pads, respectively.
The approach used to isolate NPPs in the horizontal direction is by providing an extensive common raft or diaphragm above the devices, supporting the entire nuclear island. The provision of a common mat to isolate the entire nuclear island has two main objectives: (1) to minimise relative displacements between different units of equipment; hence, to reduce the complexity of piping/flexible couplings between them, and (2) to standardise the relative displacements between the nuclear island and the non-isolated parts of the nuclear facility which need to be connected to each other. The latter condition implies that the connecting elements must be able to overcome the seismic gap between the facilities through the design of flexible expansion joints [Citation23]. shows a schematic view of a 2D system based on elastomeric bearings. In this figure, it is possible to observe the isolators located on pedestals in order to provide enough room for maintenance/inspection. In general, they are placed uniformly under the main walls and closer to each other under the reactor vessel.
Figure 8. Schematic view of 2D SPS based on elastomeric bearings [Citation57].
![Figure 8. Schematic view of 2D SPS based on elastomeric bearings [Citation57].](/cms/asset/b559c2d9-6bdc-4da1-b248-b6093d6504b7/tnst_a_980347_f0008_oc.jpg)
As mentioned in Section 3, horizontal isolation for NPPs is based on the same principles as for conventional structures. Nevertheless, a combination of two or more types of devices may be required in order to reach higher performance levels under seismic actions. Horizontal isolation may not be enough to provide full seismic protection to a nuclear facility. Consequently, full standardised seismic design is not achievable with this approach. A mixture of 2D systems with a vertical isolation system, in order to configure either a 3D or a 2D + V system, seems to be the most likely approach for a successful deployment in the next generation of NPPs.
4.3. 3D systems
The approach most often considered by researchers, in order to provide 3D seismic isolation to NPPs, is to provide horizontal and vertical isolations in one single interface, thus protecting the entire nuclear island (see ). The basic principle in most of these systems is that the horizontal isolation device is connected in series with the vertical isolation device. Two problems arise with this approach: (1) as each unit can work separately, the assurance of simultaneous performance is critical for the expected behaviour of the system, and (2) the optimal arrangement of the devices in order to make the SPS simple and efficient [Citation56]. Additionally, as the rocking modes arise as a natural consequence of a fully isolated structure, an efficient rocking suppression system should be provided as an integrated part of the SPS. It is relevant to mention that all of the 3D SPS described below are currently considered at experimental level, as no real high-risk structure (and certainly no NPP) has been equipped with this kind of technology.
Shimada et al. [Citation55] and Suhara et al. [Citation56] reported a 3D SPS based on LRBs connected in series with vertical air springs, as shown in . Experimental results on 1/7 and 1/10 small-scaled models confirmed the independency of performance of horizontal and vertical devices. This is a convenient feature for design as the dynamic properties of each device individually remain unchanged in the presence of other devices. For air springs to perform properly, it requires an air supply, air tanks and levelling systems in order to maintain a constant horizontal level of the structure.
Another 3D SPS for potential deployment in NPPs was reported by Ogiso et al. [Citation60], which was based on LRBs in series with metallic bellows subjected to high internal pressures, as shown in . Certainly, in order to work properly, this composite device needs a gas supply system, gas tanks and compressors. Experimental results obtained on a 1/5 small-scaled model confirmed that load–displacement behaviour under simultaneous vertical and horizontal loads was the same under single loads. These results confirmed that the response of each device was not affected by the presence of the other. Regarding their reliability, the author pointed out that bellows, well known as piping expansion joints, had many real applications, being effective in reducing large displacements.
At a preliminary level, Somaki et al. [Citation61] reported a 3D SPS based on elastomeric-based bearings in series and parallel with coned disk springs as shown in . As mentioned in previous sections, coned disk springs provide a versatile alternative for the vertical isolation of structures, and their application to nuclear engineering has been widely investigated. Nevertheless, for this particular case, no comprehensive tests have been reported in order to determine the combined behaviour of this device under simultaneous vertical and horizontal loadings.
Figure 9. Sketch of a 3D SPS based on lead–rubber bearings in series and parallel with coned disk springs [Citation61].
![Figure 9. Sketch of a 3D SPS based on lead–rubber bearings in series and parallel with coned disk springs [Citation61].](/cms/asset/ec999b8c-6625-4dbe-89ad-bad93542709d/tnst_a_980347_f0009_oc.jpg)
A more complex 3D SPS was reported by Kashiwazaki et al. [Citation59]. This system was based on elastomeric-based bearings in series with hydraulic load-carrying cylinders connected to accumulator units containing compressed gas, acting as vertical isolators, as shown in . During a vertical seismic load, the seismic force is converted by the load-carrying cylinders into pressure fluctuations in the fluid. The accumulator unit, which generates the vertical restoring force to be applied to the structure, is composed of two types of tank, a first-stage (variable volume) and a second-stage (constant volume). They are connected to each other through a pipe which in turn possesses an orifice. The size of the orifice can provide different levels of vertical damping forces. Experimental results performed in a small-scaled model intended to provide seismic protection for FBR, stated remarkable features of energy dissipation: a critical damping ratio in the vertical direction of about 20% and a reduction of vertical accelerations by around 2/3. Despite these results, hydraulic devices should be examined carefully for nuclear applications. Eidinger and Kelly [Citation21] commented that hydraulic devices were poor in terms of maintenance and, therefore, they might not be considered fully suitable in the nuclear industry. More and comprehensive laboratory work should be performed in order to determine and to confirm the feasibility of hydraulic devices for nuclear deployment.
Figure 10. Sketch of a 3D SPS based on elastomeric bearings in series with vertical hydraulic isolators [Citation59].
![Figure 10. Sketch of a 3D SPS based on elastomeric bearings in series with vertical hydraulic isolators [Citation59].](/cms/asset/2d769eba-b921-40a2-aea5-d16be027b00d/tnst_a_980347_f0010_oc.jpg)
An exception to the composite devices described above, which is based on the connection in series and/or parallel of two types of devices, was reported by Kageyama et al. [Citation58], who considered the provision of horizontal and vertical isolations in one single device through a 3D air spring (see ). Experimental results performed in a 1/4 small-scaled model, confirmed that the response in one direction does not affect the response in the other direction, showing independence of responses.
It is important to highlight that all of the devices described in this section need to be combined with a rocking suppression system aimed to generate restoring forces to control vertical displacements under horizontal seismic loads. A few experimental attempts have been reported in the literature in this regard. Successful results have been reported by Kashiwazaki et al. [Citation59] and Shimada et al. [Citation55], who tested rocking suppression systems based on hydraulic devices and accumulator units; Kageyama et al. [Citation58] reported results considering a rocking prevention system based on wire cables and pulleys. More theoretical and experimental work should be carried out in order to minimise the effects of rocking modes in the seismic performance of NPPs.
Up to this point, all of the devices described can be considered as the latest attempts for reaching a suitable 3D SPS for nuclear deployment. Nevertheless, early attempts to reach the same objective were reported in the literature. Coladant [Citation9] reported a potential 3D SPS based on elastomeric-based bearings, dampers and helical steel springs; Hüffmann [Citation29] reported a full base isolation system based on steel helical springs and viscous dampers; and Staudacher [Citation31] proposed a 3D SPS based on elastomeric-based bearings in combination with stiff brittle elements, called mechanical stabilisers acting as mechanical fuses under severe seismic loads. None of these systems seem to provide the level of seismic protection required for modern NPPs as no further attempts have been reported in terms of possible deployment in the nuclear industry with any of these systems.
It is also worth mentioning that recent research has reported attempts to reach 3D SPS based on elastomeric bearings formed by thick layers of rubber, for potential NPPs in China [Citation48] and Japan [Citation49,Citation70,Citation71]. However, its applicability seems limited as experimental results showed that vertical accelerations may be even amplified when the seismic ground motion contains long-period components.
4.4. 2D + V systems
A different approach to provide 3D seismic isolation is to consider a mixture of a conventional 2D system at the foundation level plus vertical isolation of the safety-related components at an upper level [Citation28,Citation51–53]. Vertical isolation is limited only to the area in which the reactor vessel and the primary coolant system are located. They are suspended using a large slab structure which in turn is vertically isolated. The provision of the slab avoids considering vertical isolation individually for each piece of equipment, jeopardising the coolant piping system that goes through them [Citation28,Citation51]. The main advantage of this approach is that the rocking motion is considerably reduced in comparison with the systems described in Section 4.3. The reduction is due to the reduced distance between the centre of gravity of the internal components and the position of the bearings. The main drawback of using this proposal is that the internal layout of the plant needs to be changed as additional space is required to accommodate the deck and its supports [Citation51].
(a) shows a schematic view of the location of this approach within the nuclear island. Horizontal isolation is provided at the foundation level by means of elastomeric-based bearings. Vertical isolation is provided only to the reactor vessel and the primary coolant system by means of coned disk springs with steel hysteretic dampers. (b) shows a schematic view of the vertically isolated common deck. Vertical isolation devices, shown in (a), are distributed around the circumference of the reactor vessel and around the perimeter of the deck. For application in FBRs, Okamura et al. [Citation28] suggested a thickness for the deck of 2 m in order to act as a rigid diaphragm, with approximately rectangular dimensions of 32 × 12 m and supported by 20 vertical isolation devices, whose unloaded height is 2.5 m.
Figure 11. Schematic view of the 2D+V system [Citation28]: (a) location of devices; (b) common deck.
![Figure 11. Schematic view of the 2D+V system [Citation28]: (a) location of devices; (b) common deck.](/cms/asset/bdbf21e2-5fdb-414f-a75a-4f9cd03fe115/tnst_a_980347_f0011_oc.jpg)
Another system, proposed by Kostarev et al. [Citation20], used an upper raft and was intended to work only in the horizontal direction. In this system, the common deck was also connected to the containment vessel rigidly in the vertical direction. Horizontal degrees of freedom were released in order to connect high-damping viscous dampers between the common deck and the containment vessel, in order to reduce floor horizontal accelerations.
The concept described in appears to be appealing for potential nuclear applications for the following reasons:
The provision of horizontal isolation at the foundation level has been widely used in conventional structures proving high reliability. On the other hand, the provision of vertical isolation, only limited to the critical parts of the nuclear island, seems to be less cumbersome than providing vertical isolation at the foundation level. Nevertheless, the reliability of vertical isolation devices may be still a pending issue.
The arrangement of the vertical isolation devices is rather simple, having no interaction with the horizontal isolation system. Independency between isolation systems may be useful for maintenance, inspection and replacement work.
Apparently, no complex rocking suppression system is required as the rocking movement may be suppressed by its nature.
5. SPS for non-safety-related elements in nuclear power plants
Non-safety-related elements are those parts of a nuclear facility which do not belong to the NPPs’ safe shutdown equipment list. Therefore, they do not normally deal with any radioactive material and/or do not perform any primary safety function [Citation72]. Nevertheless, they still play an important role within a nuclear facility as their damage may initiate secondary hazards, loss of structural integrity and functionality and/or related economic losses. Consequently, they could jeopardise the continuous operation of the whole NPP after a severe seismic event. In this light, some research has been conducted regarding the deployment of SPS for non-critical equipment of nuclear facilities.
The types of equipment and their main features may be summarised as follows:
Liquid storage tanks. They may contain corrosive substances, flammable materials, hazardous chemicals or even water as part of the cooling system. They also play a passive safety function as cooling water and the tank's concrete walls are used as radiological barriers. It is important to ensure the structural integrity and leak-tightness of these elements under the effects of severe seismic loads. The main failures reported in this type of tank are buckling of the tank wall, failure of piping system, uplift of the anchorage system and overturning of tanks. It is relevant to note that the difficulty in dealing with these structures is due to their variable weight as the storage liquid level varies continuously. Therefore, fluid–structure interaction must be considered in the design of the structure and its potential SPS. Some adverse effects may arise when using seismic isolation for protecting liquid storage tanks: increase in the fluid sloshing height and/or damages to the water circulation system due to excessive relative displacements. However, when these effects are considered correctly in the design, seismic isolation can guarantee structural integrity and continuous operation after severe earthquakes [Citation73–75]. Finally, an optimisation of seismic isolation for these elements was reported by Park et al. [Citation76], who proposed a method to estimate the cost effectiveness of seismically isolated pool structures, including fluid–structure interaction effects.
Turbines and their housing building. In general, a turbine does not belong to the safety-related equipment list. Nevertheless, it is an extremely expensive element, very difficult to manufacture, to test, to maintain and to operate. In some NPPs, turbines may be considered as part of the auxiliary heat removal safety system. In such a case, the criticality of turbines becomes clearer. Additionally, they can also be considered as a potential source of projectiles, which could happen during severe earthquakes. These projectiles may impact safety-related elements and could be the trigger of major accidents. In addition, the turbine house should be able to provide appropriate seismic safety to its equipment inside. Even though special protection against earthquakes is not normally required for these elements, the provision of SPS may be critical when higher seismic performances are required [Citation72,Citation77].
Emergency transformers and generators. Emergency equipment, such as transformers and generators, may have a superior importance within non-safety-related elements. This equipment is seismically fragile and, therefore, its continuous operation as emergency equipment must be ensured during severe earthquakes. Research reported for this type of equipment has considered the deployment of 3D SPS, highlighting the necessity to protect these elements against vertical and horizontal seismic actions [Citation78–80].
Steam generators. In general, the main failure of reported steam generators is corrosion cracking in their pipelines. The cracks may affect generators’ performance under seismic actions. For these large and slender elements, a combination of isolators at the foundation level and energy dissipators distributed along their height could be considered to provide seismic protection [Citation81].
The equipment listed above may not be the only items which need to be seismically protected in order to reduce the overall seismic risk of NPPs. Ebisawa and Uga [Citation82] proposed a methodology to select appropriate equipment in order to judiciously apply base isolation. The base isolation can be complemented with the methodology reported by Huang et al. [Citation83], determining correct seismic demands on non-safety-related components in NPPs equipped with seismic protection technologies. In general terms, seismic safety of these elements should not be overlooked as they also define the seismic vulnerability of NPPs in their entirety.
In terms of the devices considered for seismic protection of non-safety-related equipment, the majority of studies have considered the provision of SPS through passive systems. Only one study, at a theoretical level, reported the inclusion of active seismic control to equipment of NPPs [Citation84], whereas no semi-active approaches have been considered. In general, this trend is similar to that obtained for seismic protection of the nuclear island. Passive devices reported for seismic protection of non-critical equipment are the same for the nuclear island, with one exception: the consideration of variable-friction pendulum-based devices, specifically for protection of liquid storage tanks. These devices are based on the concept of friction pendulum system, but instead of having spherical sliding surfaces, they have elliptical ones. Therefore, the oscillation frequency strongly decreases with increasing sliding displacement. Variable frequency pendulum isolators (VFPI) possess one sliding surface, whereas the double variable frequency pendulum isolator (DVFPI) has two; thus, having twice the displacement capacity in comparison with VFPI. Theoretical results reported by Panchal and Jangid [Citation73] and Soni et al. [Citation75] have demonstrated their effectiveness in providing high seismic performance to the system fluid–structure isolation system. Nevertheless, it is necessary to conduct experimental tests in order to completely validate these SPS.
6. Summary, conclusions and further research
6.1. Summary
This review of the current state-of-the-art SPS in NPPs has provided the answers to the four review questions, which have been dealt with in previous sections. The answers are summarised as follows:
Why does SPS remain as an excluded subject in the nuclear industry and which issues have prevented their complete deployment in NPPs? SPS remain as an excluded subject in the nuclear industry mainly due to a lack of both experiences in real applications and specific standards and codes to design SPS for NPPs. Despite the fact that some devices have been successfully used in traditional structures (e.g. elastomeric bearings and viscous/steel hysteretic dampers), only a few devices have been deployed in high-risk structures. In addition, the experience with vertical isolation devices is very limited even for conventional structures. As a consequence, nuclear regulatory entities are not encouraged to promote/approve projects with such technology. In such a case, licences are likely to be delayed or even rejected. On the other hand, cost and safety are the factors encouraging the deployment of SPS in the nuclear industry. The cost of SPS is negligible compared to the cost of the total project and an offset is likely to be reached. Safety is indeed increased as SPS are able to ensure full structural integrity and reactor's safe shutdown after severe earthquakes.
What sorts of seismic protection devices are being considered in nuclear engineering research? Seismic protection devices considered in nuclear engineering research are mainly intended to provide passive protection. As the concepts behind seismic protection are the same for NPPs and for conventional structures, most devices can be used in both types of structures. However, as NPPs have higher requirements for seismic performance, including isolation of the vertical direction, new devices have been designed and tested experimentally. Important research has been conducted to find vertical isolators suitable for NPPs.
What are the strategies of location of SPS within NPPs? Spatial configuration of SPS in NPPs needs to be different compared to conventional structures. Horizontal isolation provided at the foundation level may not be enough to reach acceptable seismic performances in NPPs. As the inclusion of vertical isolation is desired to reach full standardised seismic design, two approaches have been reported to accommodate vertical isolation devices into NPPs. In any case, space requirements may have important influence on the NPP's layout as vertical isolators reported in the literature can occupy significant geometric dimensions.
How may SPS help, if possible, to improve structural performance and safety of NPPs under seismic actions? There is no straightforward answer for this review question. Evidence obtained from conventional structures indicates that the use of SPS has improved structural performance and seismic safety of structures. The same outcome can be inferred in nuclear applications. The safety requirements for NPPs are more demanding than for conventional structures; therefore, the establishment of optimal strategies on seismic protection can be a critical point for potential new NPPs.
6.2. Conclusions
The review of the current state-of-the-art SPS in NPPs has led to the following conclusions:
Seismic protection technologies are mature in civil engineering to mitigate the effects of earthquake actions. The significant number of seismically protected structures around the world and their satisfactory performance confirm their value for civil structures. Although few high-risk structures equipped with SPS have been deployed so far, it is highly likely that they can be used in the nuclear engineering industry with fully standardised seismic design.
In order to reach full standardisation of NPPs’ seismic design, the traditional design philosophy of seismic protection for non-critical structures must be modified. The revised philosophy should have a direct impact on two aspects: (1) redundancy, as NPPs possess higher safety requirements and seismic performance; and (2) spatial configuration, as NPPs may require the concept of seismic protection to be considered both in the horizontal and vertical directions.
The critical requirement for deployment of SPS in NPPs is the provision of vertical isolation, and consequently, the materialisation of 3D SPS. As this is not normally provided for traditional civil structures, there is a lack of experience which has acted as a drawback for its deployment in NPPs. Nevertheless, extensive and valuable research has been carried out in recent years in order to reach feasible devices able to deal with the isolation in the vertical direction and to be suitable for nuclear applications.
For NPPs, it is more effective to provide SPS to the entire island of safety-related components by means of a common isolation raft, rather than the provision of isolation to individual components. By doing this, the relative displacements between safety-related equipment are minimised and the displacements between the nuclear island and the non-isolated parts of NPPs are standardised.
The majority of devices reported for potential use in nuclear applications are intended to provide passive isolation, for protection to both the nuclear island and non-safety-related elements individually. In general, no semi-active devices have been reported for nuclear applications and only a few attempts have considered active control techniques. Passive devices are, in general, the same as those considered for traditional civil structures, having in mind the necessity to develop new types in order to deal with the requirements stated in the design philosophy stated in point 2.
Two approaches have been reported to provide 3D seismic isolation for the nuclear island: considering either one interface or two interfaces. The majority of the research reported has considered the provision of 3D isolation in one interface, i.e. the deployment of horizontal and vertical isolations in the same interface at the foundation level. However, another approach is based on the provision of horizontal isolation at the foundation level plus vertical isolation only to the reactor vessel and primary coolant system (2D + V). It seems likely that approaches that considered only horizontal isolation (2D) may not be totally suitable for NPPs in zones of moderate-to-high seismic activity, due to requirements of safety-related equipment and the impossibility of reaching full standardisation of seismic design.
Some research has been conducted regarding seismic protection for non-safety-related equipment of NPPs. Even though non-safety-related elements are not directly related to radioactive material or they are not part of primary safety systems, it seems recommendable to provide some extent of seismic protection in order to reduce the overall seismic risk of a complete nuclear facility.
6.3. Further research
The potential topics for further research in this field are suggested as follows:
Development of possible active/semi-active applications in conjunction with passive devices in order to increase the seismic safety of NPPs. A combination of these approaches may be more efficient than only using passive devices in enhancing the seismic performance of the nuclear island. These combinations can be complementary, i.e. both types of seismic protection acting together as a primary SPS; or supplementary, i.e. only passive protection acting as a primary SPS and active/semi-active control acting as a back-up system focused only on the reactor vessel and the primary coolant system. The definition of a strategy of optimal combination of these concepts of seismic protection should be developed for future design guidelines.
Assessment of the optimal spatial configuration for 3D isolation approaches, either with one or two interfaces. It is critical to determine the efficiency of the number of interfaces used, for example, by means of seismic probabilistic risk assessment (SPRA) suitable for NPPs. In this case, failure of structural and non-structural components can be better estimated through structural response parameters (e.g. floor spectral acceleration or story drifts) rather than ground-motion parameters. Therefore, fragility curves for key safety-related elements need to be estimated using such parameters. A starting point for this assessment may be, for example, the work reported by Huang et al. [Citation42]. They proposed a SPRA for NPPs isolated in the horizontal direction using fragility curves for key secondary systems as a function of the average floor spectral acceleration. Selection of adequate demand parameters to estimate fragility curves is critical to correctly assess the seismic vulnerability of NPPs. Certainly, evaluation on the efficiency of 3D seismic protection approaches can make a substantial contribution for future design guidelines. It seems suitable to focus those studies on advanced nuclear reactors for deployment in the following decades. Examples of advanced reactors are the IRIS (International Reactor Innovative and Safety) reactor, the Westinghouse AP1000TM reactor and the ELSY (European Lead-Cooled System) reactor, among others. The IRIS is under development by joint industry and academia bodies, including 20 institutions from nine countries [Citation85]. Similarly, the Westinghouse AP1000TM is a good example of successful development of new generation of reactors with design certification from the American nuclear regulatory institution [Citation86]. Also, the ELSY is aimed at developing a competitive medium-size fast reactor for deployment in the EU region [Citation87]. The first two reactors are Generation III+, whereas the last one is Generation IV. Therefore, they possess the latest technology known by the nuclear industry. Consequently, they offer appealing alternatives to attempt to reach full standardisation of NPPs’ seismic design.
More experimental work should be carried out for those devices intended to be deployed in nuclear industry. It seems clear that some devices, such as elastomeric-based isolators and steel/viscous dampers, have demonstrated their effectiveness as seismic protection technologies as shown by many real applications. However, those devices which have been designed to deal with isolation in the vertical direction may need more evidence regarding their mechanical behaviour prior to their first deployment.
Examination of the influence of non-safety-related elements in the overall seismic risk of nuclear facilities. Despite the fact that these elements do not play any primary safety functions, their importance should not be overlooked. Failure, or unacceptable performances, may jeopardise the assurance of continuous operation to the entire facility.
Nomenclature
| ABWR | = | Advanced boiling water reactor(s) |
| ADS | = | Accelerator-driven system |
| ALMR | = | Advanced Liquid Metal Reactor(s) |
| APR1400 | = | Advanced Power Reactor 1400 |
| BWR | = | Boiling water reactor(s) |
| ELSY | = | European Lead-Cooled System |
| FBR | = | Fast breeder reactor(s) |
| HDRB | = | High-damping rubber bearing(s) |
| IRIS | = | International Reactor Innovative and Secure |
| ITER | = | International Thermonuclear Experimental Reactor |
| JHR | = | Jules Horowitz Reactor |
| KALIMER | = | Korea Advanced Liquid Metal Reactor |
| LDRB | = | Low-damping rubber bearing(s) |
| LFR | = | Lead-cooled fast reactor(s) |
| LMR | = | Liquid metal-cooled reactor(s) |
| LRB | = | Lead–rubber bearing(s) |
| LWR | = | Light water reactor(s) |
| NPP | = | Nuclear power plant |
| PRISM | = | Power Reactor Inherently Safe Module |
| PWR | = | Pressurised water reactor(s) |
| RSP/I | = | Reattori ad Elevato Contenuto di Sicurezza Passiva e/o Intrinseca |
| SAFR | = | Sodium advanced fast reactor |
| SFR | = | Sodium-cooled fast reactor |
| SPS | = | Seismic protection system(s) |
| VVER-1000 | = | Water-Water Energy Reactor 1000 |
Acknowledgements
The authors would like to thank Dr Brian Ellis, Ellis Consultant, for his constructive comments on the manuscript.
References
- Martelli A. State-of-the-art on the development and application of seismic vibration control techniques and some innovatives strengthening methods for civil and industrial structures. Proceedings of International Conference on Structural Mechanics in Reactor Technology (SMiRT 17); 2003 Aug 17–22; Prague (Czech Republic).
- Martelli A, Forni M, Clemente P. Recent worldwide application of seismic isolation and energy dissipation and conditions for their correct use. Proceedings of 15th World Conference on Earthquake Engineering; 2012 Sep 24–28; Lisbon (Portugal).
- Forni M, Poggianti A, Dusi A. Seismic isolation of nuclear power plants. Proceedings of 15th World Conference on Earthquake Engineering; 2012 Sep 24–28; Lisbon (Portugal).
- Kunar RR, Maini T. Review of seismic isolation for nuclear structures. London: Electric Power Research Institute; 1979. (Report: EPRI-NP-1220-SR).
- Tajirian FF, Kelly JM, Aiken ID. Seismic isolation for advanced nuclear power stations. Earthquake Spectra. 1990;6(2):371–401.
- Higgins JPT, Green S. Cochrane handbook for systematic reviews of interventions. Chichester: Wiley-Blackwell; 2008.
- Glasziou P, Irwig L, Bain C, Colditz G. Systematic reviews in health care: a practical guide. Cambridge: Cambridge University Press; 2001.
- Egger M, Smith GD, Altman DG. Systematic reviews in health care: meta-analysis in context. London: BMJ Books; 2001.
- Coladant C. Seismic isolation of nuclear power plants – EDF's philosophy. Nucl Eng Des. 1991;127(3):243–251.
- Kato M, Sato S, Shimomura I. Utilities/industries joint study on seismic isolation systems for LWRs: Part I. Experimental and analytical studies on seismic isolation systems. Nucl Eng Des. 1991;127(3):303–312.
- Matsumura T, Sato S, Kato M. Utilities/industries joint study on seismic isolation systems for LWRs: Part II. Observed behaviors of base-isolated general buildings under real earthquakes. Nucl Eng Des. 1991;127(3):313–328.
- Suppes GJ, Storvick T. Sustainable nuclear power. Elsevier; 2007. Chapter 12, Nuclear power plant design; p. 319–351.
- Eggenberger AJ. Commentary on US R&D programs for seismic base isolation. Nucl Eng Des. 1991;127(3):239–241.
- Gluekler EL, Bigelow CC, DeVita V, Kelly JM, Seidensticker RW, Tajirian FF. Seismic isolation development for the US advanced liquid-metal reactor program. Nucl Eng Des. 1991;127(3):295–301.
- Martelli A, Masoni P, Forni M, Indirli M, Spadoni B, Pasquale Gd, Lucarelli V, Sano T, Bonacina G, Castoldi A. ENEA activities on seismic isolation of nuclear and non-nuclear structures. Nucl Eng Des. 1991;127(3):265–272.
- Austin NM, Hattori S, Rodwell E, Womack GJ. UK contribution to CEGB-EPRI-CRIEPI program on seismic isolation. Nucl Eng Des. 1991;127(3):253–264.
- Yoo B, Lee J-H, Koo G-H, Lee H-Y, Kim J-B. Seismic base isolation technologies for Korea advanced liquid metal reactor. Nucl Eng Des. 2000;199(1–2):125–142.
- Buckle IG. New Zealand seismic base isolation concepts and their application to nuclear engineering. Nucl Eng Des. 1985;84(3):313–326.
- Guéraud R, Noël-Leroux JP, Livolant M, Michalopoulos AP. Seismic isolation using sliding-elastomer bearing pads. Nucl Eng Des. 1985;84(3):363–377.
- Kostarev VV, Petrenko AV, Vasilyev PS. A new method for essential reduction of seismic and external loads on NPP's structures, systems and components. Proceedings of 17th International Conference on Structural Mechanics in Reactor Technology (SMiRT 17); 2003 Aug 17–22; Prague (Czech Republic).
- Eidinger JM, Kelly JM. Seismic isolation for nuclear power plants: technical and non-technical aspects in decision making. Nucl. Eng. Des. 1985; 84(3):383–409.
- Chang Y-W, Kuroda T, Martelli A. Overview and summary of the first international seminar on seismic base isolation of nuclear power facilities. Nucl Eng Des. 1991;127(3):233–237.
- Malushte S, Whittaker AS. Survey of past base isolation applications in nuclear power plants and challenges to industry/regulatory acceptance. Proceedings of 18th International Conference on Structural Mechanics in Reactor Technology (SMiRT 18); 2005 Aug 7–12; Beijing (China).
- Gantenbein F, Buland P. DEMT experimental and analytical studies on seismic isolation. Nucl Eng Des. 1991;127(3):409–418.
- Lo Frano R, Forasassi G. Preliminary analysis of the structural effects due to dynamic loads of the isolated next generation lead cooled reactor. Proceedings of 20th International Conference on Structural Mechanics in Reactor Technology (SMiRT 20); 2009 Aug 9–14; Espoo (Finland).
- Lo Frano R, Forasassi G. Preliminary evaluation of the seismic response of the ELSY LFR. Nuc Eng Des. 2012;242(0):361–368.
- Kato A, Moro S, Morishita M, Fujita T, Midorikawa S. A development program of three-dimensional seismic isolation for advanced reactor system in Japan. Proceedings of 17th International Conference on Structural Mechanics in Reactor Technology (SMiRT 17); 2003 Aug 17–22; Prague (Czech Republic).
- Okamura S, Kitamura S, Takahashi K, Somaki T. Experimental study on vertical component isolation system. Proceedings of 18th International Conference on Structural Mechanics in Reactor Technology (SMiRT 18); 2005 Aug 7–12; Beijing (China).
- Hüffmann GK. Full base isolation for earthquake protection by helical springs and viscodampers. Nucl Eng Des. 1985;84(3):331–338.
- Ikonomou AS. Alexisismon isolation engineering for nuclear power plants. Nucl Eng Des. 1985;85(2):201–216.
- Staudacher K. Protection for structures in extreme earthquakes: full base isolation (3-D) by the Swiss seismafloat system. Nucl Eng Des. 1985;84(3):343–357.
- Skinner RI, Bycroft GN, McVerry GH. A practical system for isolating nuclear power plants from earthquake attack. Nucl Eng Des. 1976;36(2):287–297.
- De la Llera JC, Lüders C, Leigh P, Sady H. Analysis, testing, and implementation of seismic isolation of buildings in Chile. Earthquake Eng Struct Dyn. 2004;33(5):543–574.
- International Atomic Energy Agency. Verification of analysis methods for predicting the behaviour of seismically isolated nuclear structures. Vienna: IAEA; 2002. (Report IAEA-TECDOC-1288).
- Syed MB, Patisson L, Curtido M, Slee B, Diaz S. The challenging requirements of the ITER anti seismic bearings. Nucl Eng Des. 2014;269(0):212–216.
- Lo Frano R, Forasassi G. Isolation systems influence in the seismic loading propagation analysis applied to an innovative near term reactor. Nucl Eng Des. 2010;240(10):3539–3549.
- Perotti F, Corradi dell’Acqua L, Domaneschi M, Forni M, Poggianti A, Bianchi F, Forasassi G, Lo Frano R, Pugliese G, Carelli MD, Ahmed M, Maioli A. Seismic isolation of the IRIS NSSS building. Proceedings of 20th International Conference on Structural Mechanics in Reactor Technology (SMiRT 20); 2009 Aug 9–14; Espoo (Finland).
- Hadjian AH, Tseng WS. Issues in seismic isolation of nuclear power plants. Nucl Eng Des. 1985;84(3):433–438.
- Takahashi K, Inoue K, Kato A, Morishita M, Fujita T. A development of three-dimensional seismic isolation for advanced reactor systems in Japan – part 2. Proceedings of 18th International Conference on Structural Mechanics in Reactor Technology (SMiRT 18); 2005 Aug 7–12; Beijing (China).
- Forni M, De Grandis S. Seismic-initiated events risk mitigation in lead-cooled reactors: the SILER project. Proceedings of 15th World Conference on Earthquake Engineering; 2012 Sep 24–28; Lisbon (Portugal).
- European Commission. Proposal for design guidelines for seismically isolated nuclear plants. Luxembourg: European Commission; 1995. (Report EUR 16559).
- Huang Y-N, Whittaker AS, Luco N. Seismic performance assessment of base-isolated safety-related nuclear structures. Earthquake Eng Struct Dyn. 2010;39(13):1421–1442.
- Radeva S. Multiple-model structural control for seismic protection of nuclear power plant. Nucl Eng Des. 2010;240(4):891–898.
- Soong TT, Dargush GF. Passive energy dissipation systems in structural engineering. Chichester: Wiley; 1997.
- Lee H-P, Cho M-S, Park J-Y. Developing lead rubber bearings for seismic isolation of nuclear power plants. Proceedings of 15th World Conference on Earthquake Engineering; 2012 Sep 24–28; Lisbon (Portugal).
- Domaneschi M, Martinelli L, Perotti F. The effect of rocking excitation on the dynamic behaviour of a nuclear power plant reactor building with base isolation. Proceedings of 15th World Conference on Earthquake Engineering; 2012 Sep 24–28; Lisbon (Portugal).
- Lee H-P, Cho M-S. A study on the reduction effect for seismic isolation system of nuclear power plants. Proceedings of 15th World Conference on Earthquake Engineering; 2012 Sep 24–28; Lisbon (Portugal).
- Wang F, Wang T, Ding L. Numerical and experimental study on seismic behavior of base-isolated nuclear power plant. Proceedings of 15th World Conference on Earthquake Engineering; 2012 Sep 24–28; Lisbon (Portugal).
- Okamura S, Kamishima Y, Negishi K, Sakamoto Y, Kitamura S, Kotake S. Seismic isolation design for JSFR. J Nucl Sci Technol. 2011;48(4):688–692.
- Sato N, Furukawa S, Kuno M, Shimamoto R, Takenaka Y, Nakayama T, Kondo A. Heat–mechanics interaction behavior of lead rubber bearings for seismic base isolation under large and cyclic lateral deformation. Part 2: seismic response analysis of base isolated reactor building subjected to horizontal bi-directional earthquake motions. Proceedings of 20th International Conference on Structural Mechanics in Reactor Technology (SMiRT 20); 2009 Aug 9–14; Espoo (Finland).
- Morishita M, Kitamura S, Kamishima Y. Structure of 3-dimensional seismic isolated FBR plant with vertical component isolation system. Proceedings of 17th International Conference on Structural Mechanics in Reactor Technology (SMiRT 17); 2003 Aug 17–22; Prague (Czech Republic).
- Kitamura S, Nakatogawa T, Miyamoto A, Somaki T. Experimental study on coned disk springs for vertical seismic isolation system. Proceedings of 17th International Conference on Structural Mechanics in Reactor Technology (SMiRT 17); 2003 Aug 17–22; Prague (Czech Republic).
- Kitamura S, Morishita M. Design method of vertical component isolation system. Proceedings of 16th International Conference on Structural Mechanics in Reactor Technology (SMiRT 16); 2001 Aug 12–17; Washington, DC.
- Soda S, Komatsu Y. Hybrid response control system for nuclear power station. Proceedings of 19th International Conference on Structural Mechanics in Reactor Technology (SMiRT 19); 2007 Aug 12–17; Toronto (Canada)
- Shimada T, Suhara J, Takahashi K. Study on 3-dimensional base isolation system applying to new type power plant reactor. Part 2 (Hydraulic 3-dimensional base-isolation system). Proceedings of 18th International Conference on Structural Mechanics in Reactor Technology (SMiRT 18); 2005 Aug 7–12; Beijing (China).
- Suhara J, Tamura T, Ohta K, Okada Y, Moro S. Research on 3-D base isolation system applied to new power reactor 3-D seismic isolation device with rolling seal type air spring. Part 1. Proceedings of 17th International Conference on Structural Mechanics in Reactor Technology (SMiRT 17); 2003 Aug 17–22; Prague (Czech Republic).
- Micheli I, Cardini S, Colaiuda A, Turroni P. Investigation upon the dynamic structural response of a nuclear plant on aseismic isolating devices. Nucl Eng Des. 2004;228(1–3):319–343.
- Kageyama M, Iba T, Umeki K, Somaki T, Moro S. Development of three-dimensional base isolation system with cable reinforcing air spring. Proceedings of 17th International Conference on Structural Mechanics in Reactor Technology (SMiRT 17); 2003 Aug 17–22; Prague (Czech Republic).
- Kashiwazaki A, Shimada T, Fujiwaka T, Moro S. Study on 3-dimensional base isolation system applying to new type power plant reactor (hydraulic 3-dimensional base isolation system: no. 1). Proceedings of 17th International Conference on Structural Mechanics in Reactor Technology (SMiRT 17); 2003 Aug 17–22; Prague (Czech Republic).
- Ogiso S, Nakamura K, Suzuki M, Moro S. Development of 3D seismic isolator using metallic bellows. Proceedings of 17th International Conference on Structural Mechanics in Reactor Technology (SMiRT 17); 2003 Aug 17–22; Prague (Czech Republic).
- Somaki T, Nakatogawa T, Miyamoto A, Sugiyama K, Oyobe Y, Tamachi K. Development of 3-dimensional base isolation system for nuclear power plants. Proceedings of 16th International Conference on Structural Mechanics in Reactor Technology (SMiRT 16); 2001 Aug 12–17; Washington, DC.
- Fujita T. Seismic isolation rubber bearings for nuclear facilities. Nucl Eng Des. 1991;127(3):379–391.
- Shiojiri H. CRIEPI test program for seismic isolation of the FBR. Nucl Eng Des. 1991;127(3):393–407.
- Wolf JP, Madden PA. An assessment of the application of active control to reduce the seismic response of nuclear power plants. Nucl Eng Des. 1981;66(3):383–397.
- Varpasuo P, Räty K, Kenttälä J. A trilinear base-isolator concept for nuclear power plants. Nucl Eng Des. 1980;58(3):437–448.
- Naeim F, Kelly JM. Design of seismic isolated structures: from theory to practice. New York, NY: Wiley; 1999.
- Symans MD, Constantinou MC. Semi-active control systems for seismic protection of structures: a state-of-the-art review. Eng Struct. 1999;21(6):469–487.
- Izumi M. Recent progress and development of building vibration-control systems in Japan. Nucl Eng Des. 1991;127(3):273–280.
- Seidensticker RW. Applicability of base-isolation R&D in non-reactor facilities to a nuclear reactor plant. Nucl Eng Des. 1991;127(3):291–293.
- Aoto K, Uto N, Sakamoto Y, Ito T, Toda M, Kotake S. Design study and R&D progress on Japan sodium-cooled fast reactor. J Nucl Sci Technol. 2011;48(4):463–471.
- Kubo S, Shimakawa Y, Yamano H, Kotake S. Safety design requirements for safety systems and components of JSFR. J Nucl Sci Technol. 2011;48(4):547–555.
- Kostarev VV, Petrenko AV, Vasilyev PS. An advanced seismic analysis of an NPP powerful turbogenerator on an isolation pedestal. Nucl Eng Des. 2007;237(12–13):1315–1324.
- Panchal VR, Jangid RS. Variable friction pendulum system for seismic isolation of liquid storage tanks. Nucl Eng Des. 2008;238(6):1304–1315.
- Park J-H, Moo Koh H, Kwan Kim J. Seismic isolation of pool-type tanks for the storage of nuclear spent fuel assemblies. Nucl Eng Des. 2000;199(1–2):143–154.
- Soni DP, Mistry BB, Panchal VR. Double variable frequency pendulum isolator for seismic isolation of liquid storage tanks. Nucl Eng Des. 2011;241(3):700–713.
- Park K-S, Koh H-M, Song J. Cost-effectiveness analysis of seismically isolated pool structures for the storage of nuclear spent-fuel assemblies. Nucl Eng Des. 2004;231(3):259–270.
- Danisch R, Labes M. Aseismic design of turbine houses for nuclear power plants. Nucl Eng Des. 1976;38(3):495–501.
- Choun Y-S, Kim MK, Seo J-M. Seismic and vibration isolation of an emergency diesel generator by using a spring-viscous damper system. Proceedings of 19th International Conference on Structural Mechanics in Reactor Technology (SMiRT 19); 2007 Aug 12–17; Toronto (Canada).
- Choun Y-S, Kim MK, Ohtori Y. The use of a base isolation system for an emergency diesel generator to reduce the core damage frequency caused by a seismic event. Proceedings of 19th International Conference on Structural Mechanics in Reactor Technology (SMiRT 19); 2007 Aug 12–17; Toronto (Canada).
- Ebisawa K, Ando K, Shibata K. Progress of a research program on seismic base isolation of nuclear components. Nucl Eng Des. 2000;198(1–2):61–74.
- Bhatti MA, Ciampi V, Kelly JM, Pister KS. An earthquake isolation system for steam generators in nuclear power plants. Nucl Eng Des. 1982;73(3):229–252.
- Ebisawa K, Uga T. Evaluation methodology for seismic base isolation of nuclear equipments. Nucl Eng Des. 1993;142(2–3):319–326.
- Huang Y-N, Whittaker AS, Constantinou MC, Malushte S. Seismic demands on secondary systems in base-isolated nuclear power plants. Earthquake Eng Struct Dyn. 2007;36(12):1741–1761.
- Kobori T, Kanayama H, Kamagata S. Active seismic response control systems for nuclear power plant equipment facilities. Nucl Eng Des. 1989;111(3):351–356.
- [85] Carelli MD, Conway LE, Oriani L, Petrović B, Lombardi CV, Ricotti ME, Barroso ACO, Collado JM, Cinotti L, Todreas NE, Grgić D, Moraes MM, Boroughs RD, Ninokata H, Ingersoll DT, Oriolo F. The design and safety features of the IRIS reactor. Nucl Eng Des. 2004;230(1–3):151–167.
- Schulz TL. Westinghouse AP1000 advanced passive plant. Nucl Eng Des. 2006;236(14–16):1547–1557.
- Alemberti A, Carlsson J, Malambu E, Orden A, Struwe D, Agostini P, Monti S. European lead fast reactor – ELSY. Nucl Eng Des. 2011;241(9):3470–3480.