OECD/NEA/CSNI State of the Art Report on Scaling in System Thermal Hydraulics Applications to Nuclear Reactor Safety and Design

The present paper deals with scaling in nuclear-system thermal-hydraulics (SYS TH), including the connection with Nuclear Reactor Safety Technology (NST). The paper is entirely derived from the S-SOAR document issued by CSNI of NEA, NEA/CSNI/R(2016)14, 2016 [1]. Scaling has constituted ‘an issue’ since the beginning of the exploitation of nuclear energy for civil purposes, with main reference to the generation of electricity. A Nuclear Power Plant (NPP) constitutes a technologically complex industrial system and it is characterized by the impossibility of, or the large difficulty in, characterizing the system’s performance under the conditions of the design. So, models were designed, constructed, and operated under downscaled ranges of values for one or more of the listed parameters. These features lay at the origin of the scaling issue, i.e. the difficulty in demonstrating that a model behaves like the prototype. Integrated definitions of the widely adopted terms, ‘scaling’, ‘scaling issue’, and ‘addressing the scaling issue’ are part of the present document. The related application domain includes the NST, and the licensing for water-cooled nuclear reactors under operation, under construction, or under an advanced design stage at the time of publication of the [1]. Scaling-related analyses are done in different areas of SYS TH and NST. These include the design of test facilities (both integral and separate-effect test facilities, ITF and SETF), the design of experiments (including Counterpart Test, CT), the demonstration of the capability of any computational tool, and the evaluation of uncertainty affecting the prediction of the same computational tools. A variety of approaches have been used to address the scaling issue, including non-dimensional analysis of mass, energy- and momentum-balance equations, derivation and application of scaling factors, including the hierarchy of relative importance, performing experiments at different scales, and running the SYS TH computer codes. This paper discusses the key areas and the key approach for scaling. It was found that the SYS TH computer codes, following their application to differently scaled experiments, demonstrate that the accuracy of their predictions may not depend upon the scale of the considered experiments. The TH codes also may constitute an additional valuable tool for addressing the issue of scaling.