DSMC计算中碰撞对取样和时间推进环节的高效处理方法
|
摘要
针对DSMC(Direct Simulation Monte Carlo)计算中的碰撞对取样和时间推进环节,发展了一类高效处理方法。首先引入碰撞距离的思想,发展了一种自适应碰撞距离的分子碰撞对选取方法;随后在自适应当地时间步长方法和DSMC方法数据结构特点的启发下,以模拟分子为最小时间步长调整单位,发展了一种自适应分子时间步长方法;最后以圆柱外形为例,验证了本文高效处理方法的可行性与正确性。结果表明:发展的高效处理方法能够有效放宽DSMC方法对网格尺寸的限制,显著缩短流场达到稳定所需的计算时间,并且得到满足计算精度要求的结果。
Efficient processing methods are presented to improve simulation efficiency for Direct Simulation Monte Carlo(DSMC)calculation in chains of collision pair selection and time integration.Firstly,an adaptive collision distance collision pair selection method is developed with a introduced collision distance.Then,inspired by adaptive local time step technique and data structure feature of the DSMC method,an adaptive molecule time step approach is developed,in which molecule is chosen as the smallest time step adjustment unit.Finally,the feasibility and efficacy of the efficient processing methods in this paper are verified by simulating flow over a cylinder.The results show that the developed efficient processing methods can relax the cell size constraint of the DSMC method and reduce the execution computing time significantly as well as maintain the fine simulation resolution and the flow structure.
Efficient processing methods are presented to improve simulation efficiency for Direct Simulation Monte Carlo(DSMC)calculation in chains of collision pair selection and time integration.Firstly,an adaptive collision distance collision pair selection method is developed with a introduced collision distance.Then,inspired by adaptive local time step technique and data structure feature of the DSMC method,an adaptive molecule time step approach is developed,in which molecule is chosen as the smallest time step adjustment unit.Finally,the feasibility and efficacy of the efficient processing methods in this paper are verified by simulating flow over a cylinder.The results show that the developed efficient processing methods can relax the cell size constraint of the DSMC method and reduce the execution computing time significantly as well as maintain the fine simulation resolution and the flow structure.
引文
[1]沈青.稀薄气体动力学[M].北京:国防工业出版社,2003.
[2]樊菁.稀薄气体动力学:进展与应用[J].力学进展,2013,43(2):185-201.
[3]Boyd I D.Computation of hypersonic flows using the direct simulation Monte Carlo method[C]//21st AIAA Computational Fluid Dynamics Conference,San Diego,CA,June 24-27,2013.
[4]Qu C,Wang J F.Hybrid grid DSMC method for chemical nonequilibrium with rarefied flow heating[J].Transactions of Nanjing University of Aeronautics and Astronautics,2015,32(4):409-414.
[5]Moss J N,Glass C E,Greene F A.DSMC simulation of Apollo capsule aerodynamics for hypersonic rarefied conditions[R].AIAA 2006-3577.
[6]Kim M G,Kim H S.A parallel cell based DSMC method with dynamic load balancing using unstructured adaptive meshes[R].AIAA 2003-1033.
[7]王学德.高超声速稀薄气流非结构网格DSMC及并行算法研究[D].南京:南京航空航天大学,2006.
[8]Matthias L.Optimization and parallelization of the DSMC method on unstructured grids[R].AIAA 97-2512,1997.
[9]Lebeau G J,Boyles K A,Lumpkin F E.Virtual sub-cells for the direct simulation Monte Carlo method[R].AIAA 2003-1031.
[10]黄飞,陈智,程晓丽,等.一种基于自适应碰撞距离的DSMC虚拟子网格方法[J].空气动力学学报,2014,32(4):506-510.
[11]Matthias L.Local time stepping with automatic adaptation for the DSMC method[R].AIAA-98-2670,1998.
[12]朱荣丽.过渡区DSMC数值模拟的一种自适应当地时间步长新方法[J].宇航学报,2008,29(4):1142-1146.
[13]Gallis M A,Torczynski J R,Rader D J,et al.Convergence behavior of a new DSMC algorithm[J].Journal of Computational Physics,2009,228:4532-4548.
[14]Bird G A.Molecular gas dynamics and the direct simulation of gas flow[M].Oxford:Clarendon Press,1994.
[15]陈伟芳,吴明巧,赵玉新,等.超声速平板绕流DSMC/EPSM混合算法研究[J].空气动力学学报,2002,20(4):441-445.
[2]樊菁.稀薄气体动力学:进展与应用[J].力学进展,2013,43(2):185-201.
[3]Boyd I D.Computation of hypersonic flows using the direct simulation Monte Carlo method[C]//21st AIAA Computational Fluid Dynamics Conference,San Diego,CA,June 24-27,2013.
[4]Qu C,Wang J F.Hybrid grid DSMC method for chemical nonequilibrium with rarefied flow heating[J].Transactions of Nanjing University of Aeronautics and Astronautics,2015,32(4):409-414.
[5]Moss J N,Glass C E,Greene F A.DSMC simulation of Apollo capsule aerodynamics for hypersonic rarefied conditions[R].AIAA 2006-3577.
[6]Kim M G,Kim H S.A parallel cell based DSMC method with dynamic load balancing using unstructured adaptive meshes[R].AIAA 2003-1033.
[7]王学德.高超声速稀薄气流非结构网格DSMC及并行算法研究[D].南京:南京航空航天大学,2006.
[8]Matthias L.Optimization and parallelization of the DSMC method on unstructured grids[R].AIAA 97-2512,1997.
[9]Lebeau G J,Boyles K A,Lumpkin F E.Virtual sub-cells for the direct simulation Monte Carlo method[R].AIAA 2003-1031.
[10]黄飞,陈智,程晓丽,等.一种基于自适应碰撞距离的DSMC虚拟子网格方法[J].空气动力学学报,2014,32(4):506-510.
[11]Matthias L.Local time stepping with automatic adaptation for the DSMC method[R].AIAA-98-2670,1998.
[12]朱荣丽.过渡区DSMC数值模拟的一种自适应当地时间步长新方法[J].宇航学报,2008,29(4):1142-1146.
[13]Gallis M A,Torczynski J R,Rader D J,et al.Convergence behavior of a new DSMC algorithm[J].Journal of Computational Physics,2009,228:4532-4548.
[14]Bird G A.Molecular gas dynamics and the direct simulation of gas flow[M].Oxford:Clarendon Press,1994.
[15]陈伟芳,吴明巧,赵玉新,等.超声速平板绕流DSMC/EPSM混合算法研究[J].空气动力学学报,2002,20(4):441-445.