Requirements for cfd-grade experiments for nuclear reactor thermalhydraulics - CEA - Commissariat à l’énergie atomique et aux énergies alternatives Access content directly
Conference Papers Year : 2018

Requirements for cfd-grade experiments for nuclear reactor thermalhydraulics


When looking for a validation experiment for CFD application to reactor simulation for either design or safety studies, it appears that available data often suffer from a lack of local measurements, an insufficient number of measured flow variables, a lack of well-defined initial and boundary conditions, and a lack of information on results uncertainty. Therefore the working group on CFD application to nuclear safety of the OECD-NEA-CSNI-WGAMA decided to establish some requirements for CFD-grade experiments able to validate properly the single phase CFD tools. The SILENCE network also supported this initiative and contributed to this work. This paper is a first attempt to establish such requirements. Several steps of an ideal experimental program are considered from the design to the data analysis in order to give recommendations and guidelines for future experiments. Clear objectives should be first defined with reference to a reactor application, an analysis of the process to be investigated and a selection of the type of turbulence model which may be used for the simulation. Then a discussion between experimentalists and the CFD code practitioners is necessary to define the test section geometry, the initial and boundary conditions, the quantities of interest with their locations, to define the requirements on the measurement uncertainty. The choice of measurement technique stem from these specifications. The acceptance criteria should be defined in accordance with the required accuracy and the sensitivity of the measurements to the uncertainty in the experimental conditions. Preliminary CFD simulations are necessary to confirm the most appropriate measurement locations, experimental conditions and overall the interest of the experiment for the physics to validate. Iterations may be necessary to get an optimized design of the experiment. CFD grade experiments should be able to validate CFD and one important concern is to minimize the validation uncertainty on some selected figures of merit.


Fichier principal
Vignette du fichier
201800002786.pdf (1.33 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

cea-02328974 , version 1 (04-12-2019)


  • HAL Id : cea-02328974 , version 1


D. Bestion, R. Camy, S. Gupta, C. Heib, O. Marfaing, et al.. Requirements for cfd-grade experiments for nuclear reactor thermalhydraulics. CFD4NRS : Computational Fluid Dynamics for Nuclear Reactor Safety, Sep 2019, Shanghai, China. ⟨cea-02328974⟩


184 View
288 Download


Gmail Mastodon Facebook X LinkedIn More