Development and validation of CFD methods for two-phase flow analysis

The present PhD Thesis deals with validation and development of CFD models to simulate two-phase flow in critical situations encountered in nuclear reactor power plants. Recently indeed, increasing computational capabilities and need of more detailed knowledge of two-phase phenomena taking place during reactor transients and accidents have led to attempts also in two-phase flow CFD simulations. The research activity has been carried out in the framework of two international Projects: the NURESIM (Nuclear Reactor SIMulation) Integrated Project, and the international BFBT (NUPEC BWR Full-Size Fine-mesh Bundle Tests) Void Fraction Benchmark. The reference tool adopted for the validation activity is the French research code NEPTUNE_CFD, specifically developed for multi-phase flow simulations. Two main areas of interest were identified for the selection of validation cases, concerning two of the most important needs related to reactor safety: Pressurized Thermal Shock (PTS) scenarios, and Critical Heat Flux (CHF) conditions. The starting phase of the activity, aimed at validating turbulence and interface transport models, was focused on the analyses of four separate-effects tests dealing with a very severe PTS scenario, namely the Emergency Core Cooling (ECC) Injection. During this accidental transient, the cold water injection into the uncovered Cold Leg (CL) generates complex transport phenomena in the jet region, stratified flows and phase change at the interface. Therefore, two impinging jet (1Ph and 2Ph) and two stratified flow (with and without condensation) experiments have been simulated. The following research phase aimed at validating the more complex wall boiling model; accordingly, one of the integral tests of the BFBT Void Fraction Benchmark was selected as validation case and simulated. Actually these experiments reproduce the current boiling flow configuration in the Fuel Assembly (FA) of a Boiling Water Reactor (BWR), and the microscopic scale of measured data is suitable for CFD code validation. An extensive sensitivity analysis was conducted for each validation case. The effects of some fundamental problem parameters (such as grid refinement, boundary conditions and modelling approach) was studies, in accomplishment with the Best Practice Guidelines (BPG) for the use of CFD codes in Nuclear Reactor Safety (NRS) applications. Obtained results were evaluated against experimental data and, within two of the considered cases, comparison with predictions of different CFD codes was also possible. Concluding, the analyses carried on within the present PhD Thesis contributed to the validation and development of two-phase models implemented in the NEPTUNE_CFD code, generally confirming the improvements achieved by the current CFD codes for two-phase applications.