论文部分内容阅读
复杂外形的黏性网格生成已经成为计算流体力学(CFD)在工程应用中的主要瓶颈,高效而鲁棒的网格生成技术研究显得尤为迫切.使用基于物面外形自适应加密的笛卡儿网格方法生成计算网格,投影方法拟合壁面边界,将投影得到的柱形单元沿法向分层获得适于黏性计算的网格.应用特征恢复技术恢复凹面特征并改善凸面处的网格质量.流场解算中应用最小二乘法重构获得2阶精度,Roe格式计算无黏通量,Holmes-Connell方法计算黏性通量.加入Spalart-Allmaras一方程湍流模型以获得湍流解.上下对称高斯-赛德尔迭代(LU-SGS)格式隐式时间推进,对网格重排序保证格式的稳定性和平衡性.算例显示网格生成方法是快速的和鲁棒的,流场计算精度满足工程需求,因此该方法非常适合应用于实际工程计算.“,”The difficulty of applied computational fluid dynamics (CFD) is grid generation of complex configurations. Study of quick and robust grid generation methods is therefore urgently needed. Computational grids are generated by the Cartesian grid method in this article and adaptive refinement is used at high curvature positions. Wall boundaries are snapped by the projection method, and the viscous boundary layer grids are obtained through partitioning the prism grids. Feature recovery technique is used for concave and convex surfaces. The flow solver is second order by using the least squares reconstruction technique. Roe scheme is used for inviscid flux and Holmes-Connell method is used for viscous flux. Spalart-Allmaras one equation turbulence model is used for turbulence computation. Lower-upper symmetric Gauss-Seidel (LU-SGS) scheme is used for implicit time marching. Grids are reordered for stability. The samples demonstrate that this grid generation method is robust and effective, and satisfies the engineering requirements of flow field computation. The method used in this article is suitable for engineering application.