稀疏化动态模态分解算法在发动机缸内流场研究中的应用  

Application of Sparsity DMD Method in the Research of In-Cylinder Flow Field

作　　者：秦文瑾 汪涛 齐观超 刘大明[2] 周磊[3] Qin Wenjin;Wang Tao;Qi Guanchao;Liu Daming;Zhou Lei(School of Mechanical Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China;School of Automobile and Transportation,Tianjin University of Technology and Education,Tianjin 300222,China;State Key Laboratory of Engines,Tianjin University,Tianjin 300072,China)

机构地区：[1]上海理工大学机械工程学院,上海200093 [2]天津职业技术师范大学汽车与交通学院,天津300222 [3]天津大学内燃机燃烧学国家重点实验室,天津300072

出　　处：《内燃机学报》2020年第5期426-432,共7页Transactions of Csice

基　　金：国家自然科学基金资助项目(51506118,51706155);重大研究计划培育基金资助项目(91741119);天津市教委科研计划资助项目(2017KJ114).

摘　　要：采用大涡数值模拟方法模拟了发动机缸内冷态流场,连续计算100个周期,获得了缸内多循环流场数据库,模拟结果通过粒子图像速度场测量技术(PIV)测量试验进行了验证.然后,采用动态模式分解(DMD)算法分析动态非线性系统流场数据库,以识别其流动特性.结果表明:DMD算法能够有效识别缸内涡团脉动频率,提取对应的流场结构,有利于发现在发动机整个工作过程中具有大衰变率的不稳定流场结构.此外,改进的“稀疏化”DMD算法可有效地对最重要的流场结构进行低维近似,这将有利于寻找影响和控制发动机缸内流场动态演化的方法.In-cylinder flow of an internal combustion engine in motored condition was simulated by the large eddy simulation.Flow field from 100 consecutive cycles were calculated and the simulation results were validated by particle image velocimetry(PIV)measurements.Then,dynamic mode decomposition(DMD)algorithm was used to analyze the flow field database to identify the flow characteristics.The result shows that the DMD algorithm can identify the in-cylinder vortexes vibration frequency,extract the corresponding flow structures and find the most unstable flow structures.In addition,the modified sparsity DMD algorithm is effective to select low dimensional representation for capturing the most important flow dynamic structures,which is benefit to find a way for further influencing and controlling the in-cylinder flow dynamic evolution.

关 键 词：缸内流场 大涡模拟 涡团 动态模态分解 

分 类 号：TK402[动力工程及工程热物理—动力机械及工程]