论文标题
统一气体波动粒子方法VII:旋转和振动非平衡的双原子气体
Unified gas-kinetic wave-particle methods VII: diatomic gas with rotational and vibrational nonequilibrium
论文作者
论文摘要
在近太空飞行中,围绕车辆周围的高超音速流与多尺度非平衡物理有关,其本地knudsen数量从前沿高度可压缩流到无后边缘粒子的传输。要准确捕获从连续的Navier-Stokes解决方案到单个计算中稀有气体动力学的所有流程度中的解决方案,需要真正的多尺度方法。统一的气动波颗粒(UGKWP)方法是在模拟这种多层传输的模拟中。由于波颗粒的分解,在统一的气动方案(UGK)框架下,Navier-Stokes波和动力学颗粒传输中的动力学已在系统上有效地统一。在这项研究中,基于多个温度弛豫模型开发了翻译,旋转和振动模式之间非平衡的UGKWP方法。适当捕获了不同流动方式中高速流量的实际气体效应。已经进行了数值测试,包括SOD管,正常的冲击结构,围绕二维圆柱体周围的超音速流以及围绕球体和太空车辆的三维流动,以验证UGKWP方法。与离散速度方法(DVM)基于基于粒子的直接模拟蒙特卡洛(DSMC)方法相比,UGKWP方法在计算效率,记忆降低和自动恢复多验证溶液方面显示出显着的优势。
Hypersonic flow around a vehicle in near space flight is associated with multiscale non-equilibrium physics at a large variation of local Knudsen number from the leading edge highly compressible flow to the trailing edge particle free transport. To accurately capture the solution in all flow regimes from the continuum Navier-Stokes solution to the rarefied gas dynamics in a single computation requires genuinely multiscale method. The unified gas-kinetic wave-particle (UGKWP) method targets on the simulation of such a multicale transport. Due to the wave-particle decomposition, the dynamics in the Navier-Stokes wave and kinetic particle transport has been unified systematically and efficiently under the unified gas-kinetic scheme (UGKS) framework. In this study, the UGKWP method with the non-equilibrium among translation, rotation and vibration modes, is developed based on a multiple temperature relaxation model. The real gas effect for high speed flow in different flow regimes has been properly captured. Numerical tests, including Sod tube, normal shock structure, hypersonic flow around two-dimensional cylinder and three-dimensional flow around a sphere and space vehicle, have been conducted to validate the UGKWP method. In comparison with the discrete velocity method (DVM)-based Boltzmann solver and particle-based direct simulation Monte Carlo (DSMC) method, the UGKWP method shows remarkable advantages in terms of computational efficiency, memory reduction, and automatic recovering of multiscale solution.