摘要
讨论了采用预处理方法计算低速流动时遇到的参数选择、计算格式、舍入误差等问题,提出了一种适用于粘性计算的预处理参数计算方法。计算结果表明,使用高速流的计算格式计算低速问题时,即使残值收敛,也有可能得不到正确的计算结果;AUSM+-up格式适用于低速计算;马赫数小于等于10-6时,舍入误差会造成数值解无法收敛。
Several problems in preconditioning methods are investigated numerically, including preference definition, discrete schemes and computation reliability at very low Mach numbers, and a preconditioning definition is proposed for viscous calculation. The computational results demonstrated that when computing low speed flows with schemes designed for high speed, the solution may be wrong even though the residual converges; AUSM^+-up scheme is validated reliable in low speed computations; converged solutions couldn't be obtained at very low Maeh numbers(less or equal to 10^-6) owing to round-off errors.
出处
《空气动力学学报》
EI
CSCD
北大核心
2009年第4期451-455,共5页
Acta Aerodynamica Sinica
基金
国家自然科学基金资助(项目号:90505003
10602064)
关键词
预处理
极低速流动
NS方程
AUSM+-up
舍入误差
preconditioning ' very low speed flows' Navier-Stokes equations ' AUSM^+-up
round-off erfor