Pierre Coste

A Two-Phase CFD Approach to the PTS Problem Evaluated on COSI Experiment

The study is done in the context of Pressurized Thermal Shock (PTS) CFD investigations related to PWR life duration safety studies. In the simulations of such situations direct contact condensation on free surfaces much larger than cells size is a key phenomenon. Those large interfaces require a special full set of models which has been recently implemented in the NEPTUNE_CFD code. Out of large interfaces regions, some dispersed or churn flows can take place. Both situations can be taken into account in the computational domain. The approach includes an interface recognition but not an interface reconstruction. It is evaluated on COSI experiment. COSI facility is scaled 1/100 for volume and power from a 900 MW PWR under LOCA thermal hydraulic conditions. Measurements include temperature profiles at various axial positions in the pipe. The paper focuses on one test corresponding to a situation with a low water level in the leg. It is a demanding case for two-phase CFD because -a- the liquid layer which is of prime importance in PTS studies can be quite thin in comparison with the whole computational domain -b- the emergency core cooling jet plays an important role. Results are within a reasonable range even with a coarse mesh. Calculations with finer meshes quantify the effect of a better simulation of the jet fall and impact on the wall.Copyright

A Three-Dimensional Two-Fluid Modeling of Stratified Flow with Condensation for Pressurized Thermal Shock Investigations

Abstract A three-dimensional (3-D) two-fluid model for a turbulent stratified flow with and without condensation is presented, in view of investigating pressurized thermal shock (PTS) scenarios when a stratified two-phase flow takes place in the cold legs of a pressurized water reactor. A modified turbulent K-ε model is proposed with turbulence production induced by interfacial friction. A model of interfacial friction based on an interfacial sublayer concept and three interfacial heat transfer models—namely, a model based on the small eddies–controlled surface renewal concept, a model based on the asymptotic behavior of the eddy viscosity, and a model based on the interfacial sublayer concept—are implemented into a preliminary version of the NEPTUNE code based on the 3-D module of the CATHARE code. As a first step, the models are evaluated by comparison of calculated profiles of velocity, turbulent kinetic energy, and turbulent shear stress with data in a turbulent air-water stratified flow in a rectangular channel and with data for a water jet impacting the free surface of a water pool. Then, a turbulent steam-water stratified flow with condensation is calculated, and some first conclusions are drawn on the interfacial heat transfer modeling and on the applicability of the model to PTS investigations.

Analysis of Severe Accident Scenarios and Proposals for Safety Improvements for ADS Transmuters with Dedicated Fuel

Abstract So-called dedicated fuels will be utilized to obtain maximum transmutation and incineration rates of minor actinides (MAs) in accelerator-driven systems (ADSs). These fuels are characterized by a high-MA content and the lack of the classical fertile materials such as 238U or 232Th. Dedicated fuels still have to be developed; however, programs are under way for their fabrication, irradiation, and testing. In Europe, mainly the oxide route is investigated and developed. A dedicated core will contain multiple “critical” fuel masses, resulting in a certain recriticality potential under core degradation conditions. The use of dedicated fuels may also lead to strong deterioration of the safety parameters of the reactor core, such as, e.g., the void worth, Doppler or the kinetics quantities, neutron generation time, and βeff. Critical reactors with this kind of fuel might encounter safety problems, especially under severe accident conditions. For ADSs, it is assumed that because of the subcriticality of the system, the poor safety features of such fuels could be coped with. Analyses reveal some safety problems for ADSs with dedicated fuels. Additional inherent and passive safety measures are proposed to achieve the required safety level. A safety strategy along the lines of a defense approach is presented where these measures can be integrated. The ultimate goal of these measures is to eliminate any mechanistic severe accident scenario and the potential for energetics.

Large Eddy Simulation of Highly Fluctuational Temperature and Velocity Fields Observed in a Mixing-Tee Experiment

Abstract The WATLON water experiment about fluid mixing in a tee pipe is calculated with a finite element volume method and a large eddy simulation (LES) approach, with the TRIO-U code. Its unstructured tetrahedron grids do not lead to the same noteworthy disagreements previously mentioned with Cartesian grids. Branch and main pipe inlet velocity fluctuations due to turbulence are simulated with the use of “periodic boxes.” These more realistic inlet fluctuations allow physical instabilities to develop, improving the predictions. When an elbow is added upstream of the injection, the influence of the secondary flow on temperature-averaged values and fluctuations is underlined.

Two-Phase Flow Simulations for PTS Investigation by Means of Neptune_CFD Code

Two-dimensional axisymmetric simulations of pressurized thermal shock (PTS) phenomena through Neptune_CFD module are presented aiming at two-phase models validation against experimental data. Because of PTS complexity, only some thermal-hydraulic aspects were considered. Two different flow configurations were studied, occurring when emergency core cooling (ECC) water is injected in an uncovered cold leg of a pressurized water reactor (PWR)—a plunging water jet entering a free surface, and a stratified steam-water flow. Some standard and new implemented models were tested: modified turbulent 𝑘-𝜀 models with turbulence production induced by interfacial friction, models for the drag coefficient, and interfacial heat transfer models. Quite good agreement with experimental data was achieved with best performing models for both test cases, even if a further improvement in phase change modelling would be suitable for nuclear technology applications.

Preliminary Applications of the New NEPTUNE Two-Phase CFD Solver to Pressurized Thermal Shock Investigations

The objective of this communication is to present some preliminary applications to pressurized thermal shock (PTS) investigations of the CFD two-phase flow solver of the new NEPTUNE thermal-hydraulics platform, being jointly developed by EDF (Electricite De France) and CEA (Commissariat a l’Energie Atomique) and also supported by IRSN (Institut de Radioprotection et Surete Nucleaire) and FRAMATOME-ANP. In the framework of plant life extension, the Reactor Pressure Vessel (RPV) integrity is a major concern, and an important part of RPV integrity assessment is related to PTS analysis. In the case where the cold legs are partially filled with steam, it becomes a two-phase problem and new important effects occur, such as condensation due to the Emergency Core Cooling (ECC) injections of sub-cooled water. Thus, an advanced prediction of RPV thermal loading during these transients requires sophisticated two-phase, local scale, 3D codes. In that purpose, a program has been set up to extend the capabilities of the NEPTUNE two-phase CFD solver. The major challenge is to develop new physical models to take into account the complex two-phase phenomena: free surface, jet of cold water through steam, entrainment of droplets by steam, entrainment of bubbles by water, and, above all direct contact condensation. Turbulent mixing in the liquid layer plays a central role by controlling at the same time the thermal mixing in the water and the condensation rate at the interface between liquid and vapor. A sustained effort in this area has started both at CEA and EDF. New models are developed and validated with the help of available experimental data sets, which are unfortunately very few. Beyond analytical single-effect cases, it is important to compute instrumented cases close to the life-size one, including major physical effects. This is the case of the COSI experimental facility, representing a cold leg with ECC injection, where temperature profiles have been measured. Numerical results on a steady-state COSI test are presented and analyzed, showing encouraging results. However, research work is still required to achieve a better modeling of the heat transfer and turbulence at the steam-water interface.Copyright