High-Order-Accurate Methods for Complex Unsteady Subsonic Flows
High-Order-Accurate Methods for Complex Unsteady Subsonic Flows
复制标题
DOI：
10.2514/2.591
复制
发表时间：
1999
期刊：
影响因子：
2.5
通讯作者：
M. Visbal;D. Gaitonde
中科院分区：
文献类型：
--
作者：
M. Visbal;D. Gaitonde
研究了高阶差分格式在全Navier-Stokes方程数值模拟中的应用问题。该计划利用一个隐式的，近似因式分解的时间积分方法加上空间四阶和六阶紧致差分公式和高达十阶的过滤策略。对于这最后一个方面，开发了一种一致的优化方法来处理边界附近的点，从而使精度降低到最小。调查的问题表现出许多具有挑战性的功能，实际流动，包括几个复杂的曲线网格，粘性效应，不稳定性，和三维。高阶方法被认为是非常强大的每一个考虑的问题。该算法被证明是高度准确的二阶和迎风偏置的方法相比。对于几种情况，特别是很低马赫数流动，滤波被确定为标量阻尼的上级替代方案
Several issues related to the application of very high-order schemes for the finite difference simulation of the full Navier-Stokes equations are investigated. The schemes utilize an implicit, approximately factored time-integration method coupled with spatial fourth- and sixth-order compact-difference formulations and a filtering strategy of up to tenth order. For this last aspect a consistent optimization approach is developed to treat points near the boundary resulting in minimal degradation of accuracy. The problems investigated exhibit many of the challenging features of practical flows and include several with complications introduced by curvilinear meshes, viscous effects, unsteadiness, and three-dimensionality. The high-order method is observed to be very robust for every problem considered. The algorithm is demonstrated to be highly accurate compared to both second-order and upwind-biased methods. For several cases, particularly very-low-Mach-number flows, filtering is determined to be a superior alternative to scalar damping