Abstract Turbulent mixing of warm and cold water streams within a vertical T-junction configuration of a nuclear power plant piping system causes near-wall temperature fluctuations which may lead to High-Cycle Thermal Fatigue (HCTF) failure of the wall material. To predict frequency and amplitude of such temperature fluctuations accurately the method of Large-Eddy Simulation (LES) in the numerical simulation code OpenFOAM is used. As an experimental test case, the turbulent mixing experiment of Kamide et al. [1], where a vertical branch pipe with a cold water stream is connected from below to a horizontal main pipe with a warm water stream to form a vertical T-junction configuration. From the various observed flow patterns, the temperature and velocity data of the ‘wall jet’ are chosen as a reference because it is known to have the highest potential for thermal fatigue. Incomplete mixing and an instability associated with large turbulent structures near the junction region lead to significant temperature fluctuations close to the wall before a vertical thermal stratification further downstream stabilizes the flow. The highest near-wall fluctuation amplitudes are 26% of the temperature difference between the streams. A spectral peak occurs at a Strouhal number of 0.2. The results of the simulations demonstrate, that the mean velocity and temperature profiles of the experiments are well captured, whereas the root-mean-square (RMS) temperature fluctuations deviate from the measurements in some cases, in particular when coarse grids are used. In order to find a lower limit for the required spatial resolution three different numerical meshes with a variable number of cells up to 28 × 106 are investigated. The results demonstrate that with a sufficient mesh resolution the velocity and temperature distributions as well as the spectral peak can be simulated with good agreement to the experimental data.

中文翻译：

---

垂直 T 型接头配置中湍流热流混合的大涡模拟

摘要 核电站管道系统垂直 T 型接头配置内温水和冷水流的湍流混合会导致近壁温度波动，这可能导致壁材料的高循环热疲劳 (HCTF) 失效。为了准确预测这种温度波动的频率和幅度，使用数值模拟代码 OpenFOAM 中的大涡模拟 (LES) 方法。作为实验测试案例，Kamide 等人的湍流混合实验。[1]，其中带有冷水流的垂直支管从下方连接到带有温水流的水平主管，以形成垂直 T 型接头配置。从观察到的各种流动模式，选择“壁射流”的温度和速度数据作为参考，因为已知它具有最高的热疲劳潜力。在进一步下游的垂直热分层稳定流动之前，不完全混合和与结区附近的大湍流结构相关的不稳定性导致靠近壁的显着温度波动。最高的近壁波动幅度是流之间温差的 26%。光谱峰出现在 Strouhal 数为 0.2 处。模拟结果表明，实验的平均速度和温度分布被很好地捕获，而均方根 (RMS) 温度波动在某些情况下会偏离测量值，尤其是在使用粗网格时。为了找到所需空间分辨率的下限，我们研究了三种不同的数值网格，其单元数量可变，最高可达 28 × 106。结果表明，通过足够的网格分辨率，可以模拟速度和温度分布以及光谱峰值，并且与实验数据具有良好的一致性。