• Design for plant modularisation: nuclear and SMR.

      Wrigley, Paul; Wood, Paul; Stewart, Paul; Hall, Richard; Robertson, Dan; University of Derby; Rolls-Royce Plc; University of Derby, Derby, UK; University of Derby, Derby, UK; University of Derby, Derby, UK; et al. (ASME Journals, 2018-07-22)
      The UK Small Modular Reactor (UKSMR) programme has been established to develop an SMR for the UK energy market. Developing an SMR is a multi-disciplinary technical challenge, involving nuclear physics, electrical, mechanical, design, management, safety, testing to name but a few. In 2016 Upadhyay & Jain performed a literature review on modularity in Nuclear Power. They concluded that although modularisation has been utilised in nuclear to reduce costs, more work needs to be done to “create effective modules”. Hohmann et al also concluded the same for defining modules in the chemical process plant industry. The aim of this paper is to further define modules with a particular focus on an SMR for the UK market, the UKSMR. The methods highlighted may be relevant and applied to other international SMR designs or other types of plant. An overview and examination of modularisation work in nuclear to date is provided. The different configurations are defined for the Nuclear Steam Supply System (NSSS) in primary circuits and then for Balance of Plant (BOP) modules. A top level design process has been defined to aid in the understanding of design choices for current reactors and to further assist designing balance of plant modules. The paper then highlights areas for additional research that may further support module design and definition.
    • Module layout optimization using a genetic algorithm in light water modular nuclear reactor power plants.

      Wrigley, P.A.; Wood, P.; Stewart, Paul; Robertson, D.; University of Derby; University of Sheffield; Rolls-Royce Plc (Elsevier, 2018-11-03)
      The Small Modular Reactor (SMR) concept is designed such that it will solve some of the construction problems of large reactors. SMRs are designed to be “shop fabricated and then transported as modules to the sites for installation” (IAEA, 2018). As a consequence they theoretically have shorter build schedules and require less capitalinvestment(Locatelli etal.,2014).Factory builtmodulescanalsoincreasesafetyandproductivity, dueto higher quality tools and inspection available. A literature review has highlighted substantial work has been undertaken in the research, development and construction of different types of reactors and reactor modules but the design of balance of plant modules has not been extensively researched (Wrigley et al., 2018). The focus of this paperis a casestudy for balanceofplant modulesin alightwaterreactorwhich alsocould haveapplications to other reactor types. Modules thataredesignedfor factorybuildandtransport maybebuiltinastandardized moduleapproach.By maximizing module size for transport, this maximizes work offsite, to achieve the cost and schedule savings associated. A design method needs to be developed to help support this approach. To enable this, a three step method is proposed: group components into modules, layout the modules and arrange components inside the modules. The Shearon Harris nuclear power plant was chosen for its publically available data. A previous study on this plant used matrix reordering techniques to group components and heuristically assign them to large modules, built for construction in an assembly area on site, highlighting a potential capital cost savings of 15%. This paper utilizes the same allocation of components to modules as the previous study but aims to undertake the challenge of how balance of plant modules should be arranged. The literature review highlighted that although the facility and plant layout problem has been extensively researched, mathematical layout optimization has not been applied to nuclear power plants. Many techniques for layout optimization have been developed for facilities and process plants however. The work in this paper develops an optimization model using a genetic algorithm for module layout and allocation within a nuclear power plant. This paper analysed two configurations of modules, where balance of plant modules are located on either one or two sides of the nuclear island. The objective function was to minimise pipe length. In the original research, where the plant was configured for assembly on site, the balance of plant modules are located around three sides of the nuclear island. The objective function was calculated at 14,914. As the distances are calculated rectilinearly, this number would be higher in reality as pipework has to be routed around containment. The optimization reduced the objective function by 33.9% and 37.8% for the three and four floor layouts respectively when balance of plant modules are located on two sides of the nuclear island. Furthermore, when modules are located on one side of the nuclear island, the objective function was reduced by 45.4% and 46.1% for three and four floor layouts respectively. This will reduce materials used, reduce build time and hence reduce the cost of a nuclear power plant. This method will also save design time when developing the layout of modules around the plant.
    • Module layout optimization using a genetic algorithm in light water modular nuclear reactor power plants.

      Wrigley, P.A.; Wood, Paul; Stewart, Paul; Hall, Richard; Robertson, D.; University of Derby; University of Sheffield; Rolls-Royce Plc (Elsevier, 2018-11-03)
      The Small Modular Reactor (SMR) concept is designed such that it will solve some of the construction problems of large reactors. SMRs are designed to be “shop fabricated and then transported as modules to the sites for installation” (IAEA, 2018). As a consequence they theoretically have shorter build schedules and require less capital investment (Locatelli et al., 2014). Factory built modules can also increase safety and productivity, due to higher quality tools and inspection available. A literature review has highlighted substantial work has been undertaken in the research, development and construction of different types of reactors and reactor modules but the design of balance of plant modules has not been extensively researched (Wrigley et al., 2018). The focus of this paper is a case study for balance of plant modules in a light water reactor which also could have applications to other reactor types. Modules that are designed for factory build and transport may be built in a standardized module approach. By maximizing module size for transport, this maximizes work offsite, to achieve the cost and schedule savings associated. A design method needs to be developed to help support this approach. To enable this, a three step method is proposed: group components into modules, layout the modules and arrange components inside the modules. The Shearon Harris nuclear power plant was chosen for its publically available data. A previous study on this plant used matrix reordering techniques to group components and heuristically assign them to large modules, built for construction in an assembly area on site, highlighting a potential capital cost savings of 15%. This paper utilizes the same allocation of components to modules as the previous study but aims to undertake the challenge of how balance of plant modules should be arranged. The literature review highlighted that although the facility and plant layout problem has been extensively researched, mathematical layout optimization has not been applied to nuclear power plants. Many techniques for layout optimization have been developed for facilities and process plants however. The work in this paper develops an optimization model using a genetic algorithm for module layout and allocation within a nuclear power plant. This paper analysed two configurations of modules, where balance of plant modules are located on either one or two sides of the nuclear island. The objective function was to minimise pipe length. In the original research, where the plant was configured for assembly on site, the balance of plant modules are located around three sides of the nuclear island. The objective function was calculated at 14,914. As the distances are calculated rectilinearly, this number would be higher in reality as pipework has to be routed around containment. The optimization reduced the objective function by 33.9% and 37.8% for the three and four floor layouts respectively when balance of plant modules are located on two sides of the nuclear island. Furthermore, when modules are located on one side of the nuclear island, the objective function was reduced by 45.4% and 46.1% for three and four floor layouts respectively. This will reduce materials used, reduce build time and hence reduce the cost of a nuclear power plant. This method will also save design time when developing the layout of modules around the plant.
    • Simplified and accurate stiffness of a prismatic anisotropic thin-walled box.

      Canale, Giacomo; Rubino, Felice; Weaver, Paul M.; Citarella, Roberto; Maligno, Angelo; Rolls-Royce Plc; University of Salerno; University of Bristol; University of Derby (Bentham Open, 2018-02-14)
      Background: Beam models have been proven effective in the preliminary analysis and design of aerospace structures. Accurate cross sectional stiffness constants are however needed, especially when dealing with bending, torsion and bend-twist coupling deformations. Several models have been proposed in the literature, even recently, but a lack of precision may be found when dealing with a high level of anisotropy and different lay-ups. Objective: A simplified analytical model is proposed to evaluate bending and torsional stiffness of a prismatic, anisotropic, thin-walled box. The proposed model is an extension of the model proposed by Lemanski and Weaver for the evaluation of the bend-twist coupling constant. Methods: Bending and torsional stiffness are derived analytically by using physical reasoning and by applying bending and torsional stiffness mathematic definition. Unitary deformations have been applied when evaluation forces and moments arising on the cross section. Results: Good accuracy has been obtained for structures with different geometries and lay-ups. The model has been validated with respect to finite element analysis. Numerical results are commented upon and compared with other models presented in literature. Conclusion: For cross sections with a high level of anisotropy, the accuracy of the proposed formulation is within 2% for bending stiffness and 6% for torsional stiffness. The percentage of error is further reduced for more realistic geometries and lay-ups. The proposed formulation gives accurate results for different dimensions and length rations of horizontal and vertical walls.