浅水流动的并行计算  

Parallel algorithm to analyze shallow water flow

作　　者：江春波[1] 陈立秋[1] 张庆海[1] 

机构地区：[1]清华大学水利水电工程系,北京100084

出　　处：《清华大学学报（自然科学版）》2002年第11期1548-1551,共4页Journal of Tsinghua University(Science and Technology)

基　　金：高等学校骨干教师支持计划资助项目;清华大学基础研究基金资助项目(JC2001006)

摘　　要：针对大规模水环境预测的需要,建立了网络并行机群系统(COW),进行浅水流动的并行计算研究。并行计算实现过程中引进了一种基于图论的区域分解算法,同时提出了一种针对并行计算的网格重新编号算法。作为浅水流动并行程序的验证,计算了较低Reynolds数情况下(Re≤300)圆柱绕流的流态,将各种特征物理量如Strouhal数、回流区长度等与前人的结果进行了比较。计算结果表明:浅水流动计算程序及相应的并行算法具有计算效率高、收敛速度快以及计算精度高等优点,可进一步应用于天然河道流场的并行模拟。Largescale numerical analysis of pollutant distributions in the Three Gorges Project requires a parallel processing algorithm to simulate shallow water flow. A parallel algorithm was developed using domain decomposition and a reordering algorithm. The parallel solution of the shallow water equations was verified by analyzing the flow around a circular cylinder at low Reynolds number (Re≤300) in shallow water. The results, including the values of Strouhal number and the circulation zone length, agreed well with previous results. The calculating program and its corresponding parallel algorithm are numerically stable, precise and quickly convergent. The algorithm can be used to simulate flow in rivers.

关 键 词：网络并行机群系统 计算流体力学 并行计算 浅水流动 区域分解算法 水力学 

分 类 号：TV131.4[水利工程—水力学及河流动力学]