Stato dell’arte sull’accoppiamento fra codici di sistema e di fluidodinamica computazionale.Applicazione generale su sistemi a metallo liquido pesante

The need for deeper and more accurate investigation of accidental scenarios and the challenges posed by the design of GEN IV reactors have increased the interest of the nuclear community toward CFD codes during the last years. Due to their relatively high computational costs, the CFD simulations cannot be used to replace system codes in the analysis of an entire thermal hydraulic system; they are rather meant for the analysis of local three-dimensional phenomena. A complex thermal hydraulic analysis generally requires different levels of simulations, from detailed local component-level CFD simulations to integral system-level simulations. The current state-of-the-art approaches to such multi-level analysis are mostly based on stand-alone system code and CFD simulations, even in presence of physically coupled problems (which – on the other hand – would require on-line mutual interaction and data exchange between the different solution levels). The availability of coupled simulation tools that combine system and CFD analysis would bring noticeable added value to the quality and reliability of complex thermal hydraulic studies or nuclear reactors, including liquid metal reactors. Some developments in this sense have been recently carried out by the international research and industrial community; however no well assessed coupling technology is available (or accessible) yet. The present work consists of two parts. The first part is a review of the available literature on the subject, aimed at describing the state-of-the-art as far as the development and application of system-CFD coupling methods are concerned. The second part contains the outcomes of a practical demonstrative activity, namely: an explicit coupled tool is developed and tested against a simple pipe flow problem, showing good comparisons with stand-alone simulations. The main coupling issues are identified and future works are suggested for the development of a more robust and fully-featured tool. In general, the development of a CFD-system code coupling technique is proved to be technically feasible in the frame of a well supported mid-term research program.