Band 81, Heft 3

A subchannel analysis of the steam cooling data obtained in the Rod Bundle Heat Transfer (RBHT) test facility has been performed in this study to capture the effect of spacer grids on heat transfer. The RBHT test facility has a 7 × 7 rod bundle with heater rods and with seven spacer grids equally spaced along the length of the rods. A method based on the concept of momentum and heat transport analogy has been developed for calculating the subchannel bulk mean temperature from the measured steam temperatures. Over the range of inlet Reynolds number, the local Nusselt number was found to exhibit a minimum value between the upstream and downstream spacer grids. The presence of a spacer grid not only affects the local Nusselt number downstream of the grid but also affects the local Nusselt number upstream of the next grid. A new correlation capturing the effect of Reynolds number on the local flow restructuring downstream as well as upstream of the spacer grids was proposed for the minimum Nusselt number. In addition, a new enhancement factor accounting for the effects of the upstream as well as downstream spacer grids was developed from the RBHT data. The new enhancement factor was found to compare well with the data from the ACHILLLES test facility.

The thermal hydraulic characteristics of a mixing vane grid are largely dependent on the structure of key components, such as strip, spring, dimple, weld nugget, as well as the mixing vane configuration. In this paper, several types of spacer grids with different dimple shapes are modeled under subcooled boiling conditions. Prior to the application of CFD on the dimple shape analysis, the mixing effects of spacer grids were studied. After the dimple shape analysis, the side channel effect is discussed by comparing the simulation results of a 3 × 3 and a 5 × 5 spacer grid. The two phase flow CFD models in this study are validated through simple geometry showing that the calculated void fraction is in good agreement with the experimental data. The dimple comparison result shows that varying dimple structures can result in different temperatures, lateral velocities and void fraction distributions downstream of the spacer grids. Comparison of two sizes of spacer grids demonstrate that the side channel generates different flow distribution pattern in the center channel.

During large loss-of-coolant accidents in nuclear reactors, water splatters as the quench front propagates at the quenching surface, and many droplets of different sizes and velocities are generated and carried with the steam in the cooling channel. Heat transfer due to droplets striking an overheated fuel rod above the Leidenfrost point temperature is important for predicting the peak cladding temperature. This study investigated the dynamics and heat transfer characteristics when a single droplet at room temperature collided with a surface at 425 °C experimentally, using synchronized high-speed video and infrared cameras. Various physical parameters related to heat transfer model development were measured, including the residence time, spreading diameter, local heat flux distribution, effective heat transfer area, average vapor film thickness, and total heat transfer per collision. The measured data were compared with the values of the physical parameters predicted by existing mechanistic models.

Mixing vane grids (MVG) have great influence on coolant temperature field in the rod bundle. The MVG could enhance convective heat transfer between the fuel rod wall and the coolant, and promote inter-subchannel mixing at the same time. For the influence of the MVG on convective heat transfer enhancement, many experiments have been done and several correlations have been developed based on the experimental data. However, inter-subchannel mixing promotion caused by the MVG is not well estimated in subchannel analysis because the information of mixing vanes is totally missing in most subchannel codes. This paper analyzes the influence of mixing vanes on coolant temperature distribution using the improved MVG model in subchannel analysis. The coolant temperature distributions with the MVG are analyzed, and the results show that mixing vanes lead to a more uniform temperature distribution. The performances of split vane grids under different power conditions are evaluated. The results are compared with those of spacer grids without mixing vanes and some conclusions are obtained.

Clad ballooning and the resulting partial flow blockage are one of the major thermal-hydraulic concerns associated with the coolability of partially blocked cores during a loss-of-coolant accident (LOCA). Several in-pile tests have shown that fuel relocation causes a local power accumulation and a high thermal coupling between the clad and fuel debris in the ballooned regions. However, previous experiments in the 1980s did not take into account the fuel relocation phenomena and resulting local power increase in the ballooned regions. The present paper presents the results of systematic investigations on the coolability of rod bundles with flow blockages. The experiments were mainly performed in 5 × 5 rod bundles, 2 × 2 rod bundles and other test facilities. The experiments include a reflood heat transfer, single-phase convective heat transfer, flow redistributions phenomena, and droplet break-up behavior. The effects of the fuel relocation and resulting local power increase were investigated using a 5 × 5 rod bundle. The fuel relocation phenomena increase the peak cladding temperature.

As pressure drop is one of the most critical thermal hydraulic parameters for spacer grids the accurate estimation of it is the key to the design and development of spacer grids. Most of the available correlations for pressure drop do not contain any real geometrical parameters that characterize the grid effect. The main functions for spacer grid are structural support and flow mixing. Once the boundary sublayer near the rod bundle is disturbed, the liquid forms swirls or flow separation that affect pressure drop. However, under two phase flow conditions, due to the existence of steam bubble, the complexity for spacer grid are multiplied and pressure drop calculation becomes much more challenging. The influence of the dimple location, distance of mixing vane to the nearest strip, and the effect of inter-subchannel mixing among neighboring subchannels on pressure drop and downstream flow fields are analyzed in this paper. Based on this study, more detailed space grid geometry parameters are recommended for adding into the correlation when predicting pressure drop.

Previous studies showed that the subchannel codes need most CPU time to solve the matrix formed by the conservation equations. Traditional matrix solving method such as Gaussian elimination method and Gaussian-Seidel iteration method cannot meet the requirement of the computational efficiency. Therefore, a new algorithm for solving the block penta-diagonal matrix is designed based on Stone's incomplete LU (ILU) decomposition method. In the new algorithm, the original block penta-diagonal matrix will be decomposed into a block upper triangular matrix and a lower block triangular matrix as well as a nonzero small matrix. After that, the LU algorithm is applied to solve the matrix until the convergence. In order to compare the computational efficiency, the new designed algorithm is applied to the ATHAS code in this paper. The calculation results show that more than 80 % of the total CPU time can be saved with the new designed ILU algorithm for a 324-channel PWR assembly problem, compared with the original ATHAS code.

In pressurized water reactors (PWRs), the spacer grids of the fuel assembly has significant impact on the thermal-hydraulic performance of the fuel assembly. Particularly, the spacer grids with the mixing vanes can dramatically enhance the secondary flow and have significant effect on the void distribution in the fuel assembly. In this paper, the CFD study has been carried out to analyze the effects of the spacer grid with the steel contacts, dimples and mixing vanes on the boiling two-phase flow characteristics, such as the two-phase flow field, the void distribution, and so on. Considered the influence of the boiling phase change on two-phase flow, a boiling model was proposed and applied in the CFD simulation by using the UDF (User Defined Function) method. Furthermore, in order to analyze the effects of the spacer grid with mixing vanes, the adiabatic (without boiling) two-phase flow has also been investigated as comparison with the boiling two-phase flow in the fuel assembly with spacer grids. The CFD simulation on two-phase flow in the fuel assembly with the proposed boiling model can predict the characteristics of two-phase flow better.

Currently, the prediction of rod bundle CHF is dependent on CHF correlations derived from CHF data. A simple correction factor, such as F-factor, is often used to account for the axial power shape differences based on a simple accumulated energy concept, which has totally no consideration on the impact of true local condition on CHF mechanism. Subsequently, as expected, large uncertainty is often associated with the CHF value and CHF location predictions. For the purpose of obtaining different power shapes effects on CHF, CFD calculated parameter values were used to predict the possible CHF occurrence location. The possible CHF location prediction method proposed in this paper is calculated void fraction, heat transfer coefficient (HTC), liquid temperature distribution and detailed local parameters. And the uniform and non-uniform CHF were analyzed. The prediction of possible CHF locations in a 5 × 5 rod bundle may provide useful information for the design of a full-length CHF test, enhance the accuracy of CHF and CHF location prediction, and reduce the costs of the experimentation.

The integrity and thermal hydraulic characteristics of steam generator are of great concern in the nuclear industry. The tube support plates (TSP), one of the most important components of the steam generator, not only support the heat transfer tubes, but also affect the flow dynamic and thermal hydraulic characteristics of the secondary-side flow inside the steam generator. Different working conditions, ranging from single-phase adiabatic condition to two-phase high-void boiling condition, are simulated and analyzed. Calculated void fraction, under simple geometry, agrees well with the experiment data whilst the simulated heat transfer coefficient is tremendously close to the empirical correlation. Temperature, void fraction, and velocity distributions in different locations show reasonable distribution. The simulation results indicate that TSP can enhance the heat transfer in the secondary side of the steam generator. On the top of TSP, with the increase in cross-section flow area, the back-flow phenomenon occurs, which might lead to the contamination of precipitation.

Analysis was performed for a heater infrared thermometric imaging temperature data obtained from high heat flux pool boiling and liquid film boiling experiments BETA. With the OpenFOAM solver, heat flux distribution towards the coolant was obtained by solving transient heat conduction of heater substrate given the heater surface temperature data as boundary condition. The so-obtained heat flux data was used to validate them against the state-of-art wall boiling model developed by Shaver [1] with the assumption of micro-layer hydrodynamics. Good agreement was found between the model prediction and data for conditions away from the critical heat flux (CHF). However, the data indicate a different heat transfer pattern under CHF, which is not captured by the current model. Experimental data strengthen the notion of burnout caused by the irreversible hot spot due to failure of rewetting. The observation forms a basis for a detailed modeling of micro-layer hydrodynamics under high heat flux.

Spacer grids could cause heat transfer enhancement both at the spacer grid regions and downstream of the spacer grids as a result of mixing promoted by the spacer grids in the rod bundle. This phenomenon has been demonstrated by many experiments, and several correlations have been developed based on these experimental data. This paper gives a review of the grid-enhanced heat transfer correlation developments in single phase flow. Following the exploration of the correlation development history, a predictive formulation of grid-enhanced heat transfer in single phase flow is established taking into account the effect of both swirl flow and crossflow. With emphasis on modeling of the mixing mechanism associated with the mixing vane grid, the new correlation could better reflect the physical process of the heat transfer augmentation, while a large number of experimental data are needed to determine the coefficients of the new correlation.

We evaluated the performance of various turbulence models, including eddy viscosity models and Reynolds stress models, when analyzing rod bundles in fuel assemblies using the Computational Fluid Dynamics (CFD) method. The models were assessed by calculating the pressure drop and Nusselt numbers in 5 × 5 rod bundles using the CFD software ANSYS CFX. Comparisons between the numerical and experimental results, as well as the swirl factor, cross-flow factor, and turbulence intensity utilized to evaluate the swirling and cross-flow, were used to analyze the inner relationship between the flow field and heat transfer. These comparisons allow the selection of the most appropriate turbulence model for modeling flow features and heat transfer in rod bundles.

The energetic fuel-coolant interaction (FCI) has been one of the primary safety concerns in nuclear power plants. Graphical processing unit (GPU) implementation of the moving particle semi-implicit (MPS) method is presented and used to simulate the fuel coolant interaction problem. The governing equations are discretized with the particle interaction model of MPS. Detailed implementation on single-GPU is introduced. The three-dimensional broken dam is simulated to verify the developed GPU acceleration MPS method. The proposed GPU acceleration algorithm and developed code are then used to simulate the FCI problem. As a summary of results, the developed GPU-MPS method showed a good agreement with the experimental observation and theoretical prediction.

Two-phase flow instability is a vitally important area of study for a large number of industrial systems. Density Wave Oscillation (DWO) is the most common type of flow instability caused by the change in flow rate or power in boiling systems. The code RELAP5 is used to simulate single channel, 2 × 2 subchannels, and 3 × 3 subchannels with typical BWR subchannel geometry. The onset of flow instability determinating criterion and the results of simulations are utilized to create a stable boundary. The stable boundary of a single channel is compared with those from results of other researchers. Some conclusions are made as follows. 3 × 3 subchannels are more stable than single channel and 2 × 2 subchannels. Open subchannels possess a larger stable region than close channels. The heating model is analyzed determining that asymmetrical heating has negative effect on stability as compared to symmetric heating. With the analysis of transit time, period and subcooling number, there is a positive linear relationship between the subcooling number and oscillation period.