何枫++姚朝晖++张锡文++刘正先
摘 要：對Re数不高的热对流流场，目前流场CFD数值模拟策略主要采用大涡模拟和时均湍流模式两种方法。在进行大涡数值模拟时，其中亚格子模式分别采用Smagorinsky-Lilly模式、壁面自适应当地涡粘模式、Smagorinsky-Lilly动力亚格子模式、Vreman的亚格子模式，主要对比各种亚格子模式在计算热对流时的优势和差异。进行时均湍流模式数值模拟时，采用各种湍流模式，对比计算差异，并与LES对比，确定对热对流湍流信息的抑制程度，为今后改进湍流模式奠定基础。在热对流的数值模拟中，网格技术也是关键，采用根据初算结果，调整网格密度分布技术，加快计算收敛。根据检测固定点参数的变化以及热通量平衡状态，判定计算是否收敛。实验分别采用激光粒子测速、热线进行速度测量，采用热电偶测量温度，校正CFD计算结果。
关键词：热对流 湍流 数值模拟
Progress Report of 2012
He Feng1 Yao Chaohui1 Zhang Xiwen1 Liu Zhengxian2
（1.Tsinghua University； 2.Tianjin University）
Abstract：Currently，the solution strategies for not high Re number convection is large eddy simulation method and RANS with turbulent models method. For the LES method， various sub-grid scale models are used， such as Smagorinsky- Lilly model， Lillys dynamic Smagorinsky model Vremans model and wall-adapting local eddy-viscosity model， to compare the difference and the applicability for convection flow field. The simulation data with the turbulent models are compared with the experimental data and LES data in order to determine that how much turbulent information is restrained by the method of turbulent models. Grid technique also play key role for simulation convergence. Adjusting grid density and monitoring surface variables and the balance of the heat transfer flux can help to reach the simulation convergence. Some related experiments are done by the particle image velocimetry （PIV） and hot wire technology.
Key Words：Convective flow； Turbulence； Numerical simulation
