基于分离涡模拟的核主泵全工况水力性能计算精度研究

为了研究分离涡模拟模型对核主泵水力性能预测精度的影响,在六面体结构化网格的基础上,采用基于SST κ-ω的分离涡模拟（DES）、延迟分离涡模拟（DDES）和改进的延迟分离涡模拟（IDDES）,分别进行全流量工况条件下的非定常数值计算,并与RNG κ-ε模型的计算结果作对比,从相对计算误差的大小和离散度2个方面对4种湍流模型的计算精度进行综合评判。研究结果表明：各分离涡模拟模型在全流量工况点的综合计算精度远高于RNG κ-ε模型;DDES和IDDES模型在全流量工况条件下的计算精度基本一致,并且都高于DES模型;DDES模型在设计工况点附近的计算精度明显优于IDDES。综合比较来看,DDES模型更适用于核主泵的性能预测。     

光CVD系统气流模式分析

本文对光ＣＶＤ反应系统的气流模式进行了分析，讨化了４种不同的气流动状态，计算出附面层线性底层厚度为０。１６３ｍｍ，提出用保护气体隔离工作气体，以防止石英窗内侧面淀积薄膜的可行性。     

Thermal Comfort in the Near-Zone of a Radiator Air Device

Abstract The near-zone of an air inlet device is the most critical part of a room as regards fulfilment of the thermal conform demands. An acceptable solution for residential buildings can be to supply the outdoor air through an air device placed behind a radiator. Such a device provides a mechanical exhaust system that supplies pre-heated outdoor air. Normally radiator air devices are located below a window and a window-sill. To examine the influence of a window-sill placed above an air device, several different positions and breadths of sills were tested. Air velocities and air temperatures were measured at distances within 0.50 m from the device. The measurements show that thermal comfort is strongly dependent on the location of the window-sill in relation to the air device. The degree of turbulence intensity affects the percentage of dissatisfied people due to draught. The velocity measurement system used in this study allows variation of the time constant of the system. When different time constants were used, the measured turbulence intensity varied considerably. Some measurements were carried out with the radiator switched off. In these cases different air velocities were measured, depending on the orientation of the sensors. This seems to be due to the temperature and velocity gradients at the measuring points.     

Flow Measurement Using Flying ADV Probes

An innovative measurement system of “flying” acoustic Doppler velocimeters was designed in order to allow rapid velocity measurement over a large flow field. Such measurements are necessary, for example, when measuring over a temporally varying and locally nonuniform rough bed. The measurement technique was verified by comparison with measurements taken in the same flows using a traditional stationary probe technique. Comparison showed that the flying-probe approach performs similarly to stationary measurements in capturing the mean flow field and turbulent fluctuations. The data obtained from flying probe experiments can be used to describe the flow in terms of double-averaged hydrodynamic variables, obtained by averaging in time and spatial domains within a thin slab parallel to the mean bed. Examples are presented of flow measurements over a fixed flat bed, a fixed dune bed, and over mobile developing bed forms. It is shown that near-bed measurements suffer from boundary reflection interference, though affected data can be filtered out based on the ADV-measured correlation coefficient. Measurement below roughness tops is possible, with in-bed records being detectable by spikes in measured signal-to-noise-ratio and by comparison with measured bed topography.     

Purging of a Neutrally Buoyant or a Dense Miscible Contaminant from a Rectangular Cavity. II: Case of an Incoming Fully Turbulent Overflow

Fully three-dimensional (3D) large-eddy simulation calculations of the flow past two-dimensional cavities for the case in which the incoming flow is fully turbulent are conducted to study the purging of neutrally buoyant or dense miscible contaminants introduced instantaneously inside the cavity. 3D simulations are needed because in the turbulent case (TC), as opposed to the laminar inflow case (LC) considered in the companion paper, the interactions between the coherent structures advected from the incoming channel and the eddies inside the cavity are highly 3D and have a nonnegligible effect on the mass exchange processes between the cavity and channel. Similar to the LC, it is found that the mechanism of removal of the contaminant is very different between the neutrally buoyant and buoyant cases. In the neutrally buoyant TC simulation the contaminant is ejected from the cavity due to the interactions among the large scale eddies in the separated shear layer, the coherent structures convected from the upstream channel over the cavity, and the main recirculation eddies inside the cavity. In the TC simulation with a negatively buoyant contaminant, internal wave breaking is observed to occur over the initial phases of the mixing which, along with other turbulent mixing phenomena, reduces the mean density gradient across the density interface. In the later stages, the contaminant removal and mixing processes are controlled by the interactions of the trailing edge vortex with the bottom layer containing denser contaminant beneath it and upstream of it (for the final stages when the vortex touches the cavity bottom). The oscillations in the size, position, and intensity of the trailing edge vortex are larger than the ones observed in the LC. As expected, turbulent mixing accelerates the purging process in the TC simulations.     

旋翼翼型动态失速特性的数值仿真研究

旋翼翼型动态失速具有很强的非定常性与非线性,为研究动态失速发生时的气动和流场特性,基于CFD软件FLU-ENT,采用单方程S-A湍流模型,对NACA0012翼型的动态失速特性进行数值仿真。仿真结果显示,轻失速是后缘分离,旋涡影响范围小;深失速则形成了很强的前缘分离涡,结构复杂,影响范围大。参数研究显示,较大的湍流强度能减缓动态失速的发生,并且动态失速状态下的分离涡影响范围较小。通过与实验的对比证明,用S-A湍流模型的方法可以较好地仿真动态失速过程中旋涡的形成与发展过程。     

Modelling the gas exchange rate in perforation-mediated modified atmosphere packaging: Effect of the external air movement and tube dimensions

Perforation-mediated modified atmosphere packages control the gas exchange rate and thus the internal atmosphere composition by judicious choice of dimensions of small tubes in perforations in an otherwise hermetic container. The aims of this work were (i) to study the effect of external turbulence on the gas exchange rate through small perforations, (ii) to develop a mathematical model to predict the effect of tube dimensions and hydrodynamic conditions on gas exchange coefficients, and (iii) validate the model developed during the storage of shredded carrots. Different hydrodynamic conditions in the controlled temperature room due to the storage temperature settings (5, 10 and 15 °C) and PVC tubes dimensions (diameter from 1.5 to 4.5 mm length from 2.0 to 6.0 mm) were tested. Results showed that, as expected, hydrodynamic conditions (induced turbulence) and tube diameter had a significantly positive effect on the mass transfer values, whereas tube length had a negative effect. The permeability ratio between CO₂ and O₂ was 0.83 ± 0.01 and did not show any pattern with the factors tested. The gas exchange coefficients increased with decreasing temperature, as the establishment of lower temperatures implied a frequent air movement. A mathematical model to estimate changes of gas composition over time as a function of tube dimensions and hydrodynamic conditions was developed and found to have good predictive abilities. The model was validated during the storage of shredded carrots at 5 °C in PM-MAP, showing a good agreement between the experimental and predicted gas exchange of O₂ and CO₂. The results obtained in this work showed that hydrodynamic conditions of the cold storage room affected the gas exchange rate in PM-MAP.     

Application of Proper Orthogonal Decomposition Method in Wind Field Simulation for Roof Structures

The weighted-amplitude wave superposition method and the proper orthogonal decomposition (POD) technique are first introduced. Then the POD technique combined with a wind field reconstruction method for a flat roof is studied. In the meantime, wind field extension using B-spline interpolation and the wind field reconstruction are verified. On this basis, a three-step simulation algorithm to generate wind field is presented for roof structures. In the proposed algorithm, the weighted-amplitude wave superposition method, the POD technique, and the B-spline interpolation technique are combined together. In contrast to the existing algorithms in frequency-domain-based simulation, the proposed one can reduce the computation cost remarkably. In order to demonstrate the capability and efficiency of the proposed algorithm, an engineering example, the wind field simulation for a large-scale roof of the Foshan Century Lotus Stadium, is illustrated. The results show that the proposed algorithm is very efficient with sufficient precision. Therefore, it is advantageous in wind resistance analysis for large-scale roofs.     

Rapid Profiling of an Evolving Bed Form Using Planar Laser Sheet Illumination

Traditional methods of measuring the profile of a scour hole or bed form have poor temporal resolution and may require the temporary cessation of the flow in order to be executed. These are undesirable characteristics since many hydraulic flows have unsteady water-sediment interfaces that can display considerable differences between their “dynamic” (flow on) and “static” (flow off) states. The present technical note discusses the application of planar laser sheet illumination to the erosion of a cohesionless granular bed due to a planar turbulent wall jet. The measurements allow for the quantitative study of the evolution of the bed in its earliest stages. In some cases an ephemeral bed form is observed to form prior to the development of the main bed form. Additionally, the measurements illustrate the oscillatory nature of the bed form under certain conditions. The experiments therefore demonstrate the great potential of this experimental technique for gaining previously inaccessible information on scour processes.     

Modeling of the fluctuating component in the form of the sum of an infinite number of random quantities. Part 2. Model of transfer of turbulent stresses and turbulent heat fluxes

The paper presents an approach to construction of turbulence models that allows modeling of the fluctuating component in the form of the sum of an infinite number of random quantities. The first part of the paper deals with the technique of construction of the k-ε type model. In the second part of the paper, construction of the model of transfer of turbulent stresses and heat fluxes is considered. It is shown that the technique does not depend on the choice of the dissipative variable.