首页> 外文会议>AIAA SciTech forum and exposition >Wall-modeled LES of the NASA Juncture Flow Experiment
【24h】

Wall-modeled LES of the NASA Juncture Flow Experiment

机译:NASA接合流实验的墙面模型LES

获取原文

摘要

We perform wall-modeled large eddy simulation (WMLES) of the NASA Juncture Flow experiment (Kegerise & Neuhart, NASA/TM-2019-20286) using an equilibrium wall model and unstructured finite volume solver, to assess its predictive ability for complex geometries. The flow condition simulated corresponds to 5 degrees angle of attack with a freestream Mach number of 0.189, and a Reynolds number based on the mean crank chord of 2.4 million. AH simulations neglect the effects of wind tunnel walls, sting and mast present in the experiment. To assess the performance of WMLES on simpler flows with very coarse grids, we simulate flow in a turbulent channel at a friction Reynolds number, Re_τ ≈ 2000, and flow through a square duct at an Re_τ ≈ 1000. The duct flow results using the equilibrium wall model indicate that the skin friction is inaccurate near the corner for grid topologies and resolutions typically used in WMLES, having implications for more complicated flows such as the juncture flow. For the juncture flow simulation, two different approaches are investigated. The first approach consists of a truncated-domain simulation wherein the inflow plane of the computational domain is placed at ≈ 0.2 chord length of the wing, and we prescribe the mean flow from a separate Reynolds-averaged Navier-Stokes (RANS) solution along with synthetic turbulence to initiate realistic unsteadiness in the domain. The second approach involves simulating the entire geometry with trips to trigger transition to turbulence. The truncated- and full-domain simulations contain about 62 and 90 million cells, respectively; with 8-10 points per boundary layer thickness in the wall-parallel directions, 14-20 points in the wall-normal direction and a near-wall viscous spacing (Δn_1~+) of ≈ 100. Preliminary results are encouraging overall in terms of the prediction of wall pressure, wall skin friction, velocity and stresses; but indicate that further work is required to improve the predictions in the separation bubble and wing-fuselage corner regions. The grid resolution used in this study is still fairly coarse, and the results should be interpreted as work in progress.
机译:我们使用平衡壁模型和非结构化有限体积求解器对NASA接合流实验(Kegerise&Neuhart,NASA / TM-2019-20286)进行壁建模的大涡模拟(WMLES),以评估其对复杂几何形状的预测能力。模拟的流动条件对应于5度迎角,自由马赫数为0.189,基于平均曲柄弦的雷诺数为240万。 AH模拟忽略了实验中存在的风洞壁,刺痛和桅杆的影响。为了评估WMLES在具有非常粗糙的网格的简单流上的性能,我们模拟了在具有雷诺数Re_τ≈2000的湍流通道中的流动,并在Re_τ≈1000的情况下流经方管。墙模型表明,对于WMLES中通常使用的网格拓扑和分辨率,角部附近的皮肤摩擦是不准确的,这对更复杂的流(例如接合点流)有影响。对于接合点流动仿真,研究了两种不同的方法。第一种方法包括截断域仿真,其中计算域的流入平面位于机翼的≈0.2弦长处,我们规定了来自单独的雷诺平均Navier-Stokes(RANS)解决方案的平均流量以及合成湍流引发领域内的现实不稳定。第二种方法涉及通过跳动来模拟整个几何形状,以触发向湍流的过渡。截断域仿真和全域仿真分别包含大约62个和9000万个单元。在壁平行方向上每个边界层厚度为8-10点,在壁垂直方向上为14-20点,并且近壁粘性间距(Δn_1〜+)约为100。初步结果在总体上令人鼓舞预测壁面压力,壁面摩擦力,速度和应力;但指出需要做进一步的工作来改善对分离气泡和机翼机身角部区域的预测。本研究中使用的网格分辨率仍然相当粗糙,结果应解释为正在进行的工作。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号