• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2018, Vol. 54 ›› Issue (18): 146-153.doi: 10.3901/JME.2018.18.146

• 运载工程 • 上一篇    下一篇

液力变矩器热流动尺度解析模拟与特性精确预测

刘春宝1,2,3, 李静1, 徐志轩1, 马文星1,2   

  1. 1. 吉林大学机械科学与工程学院 长春 130022;
    2. 吉林大学汽车模拟与控制国家重点实验室 长春 130022;
    3. 吉林大学工程仿生教育部重点实验室 长春 130022
  • 收稿日期:2017-10-08 修回日期:2018-06-12 出版日期:2018-09-20 发布日期:2018-09-20
  • 通讯作者: 刘春宝(通信作者),男,1979年出生,博士,教授,博士研究生导师。主要研究方向为流体驱动与仿生机器人、液力传动以及流体机械流动数值模拟。E-mail:liuchunbao@jlu.edu.cn
  • 作者简介:李静,男,1993年出生,硕士研究生。主要研究方向为液力传动。E-mail:lij16@mails.jlu.edu.cn;徐志轩,男,1990年出生,硕士研究生。主要研究方向为液力传动。E-mail:xzxxde@126.com;马文星,男,1963年出生,博士,教授,博士研究生导师。主要研究方向为液力传动与自动变速。E-mail:mawx@jlu.edu.cn
  • 基金资助:
    吉林省重点科技攻关(20170204066GX)、吉林省教育厅“十三五”科学技术(JJKH20180137KJ)和工信部高端土方机械绿色设计平台建设([2017]327)资助项目。

Scale-resolving Simulation of Thermal Flow and Accurate Performance Prediction in Hydrodynamic Torque Converter

LIU Chunbao1,2,3, LI Jing1, XU Zhixuan1, MA Wenxing1,2   

  1. 1. School of Mechanical Science and Engineering, Jilin University, Changchun 130022;
    2. State Key Laboratory of Automotive Simulation and Control, Jilin University, Changchun 130022;
    3. Key Laboratory of Bionic Engineering, Ministry of Education, Jilin University, Changchun 130022
  • Received:2017-10-08 Revised:2018-06-12 Online:2018-09-20 Published:2018-09-20

摘要: 针对液力变矩器数值模拟中雷诺时均方法(RANS)占据主导地位,但预测精度较低且缺乏对流场结构准确描述的现状,利用尺度解析模拟(Scale-resolving simulation,SRS)方法实现其原始特性及内流场的模拟分析。结果表明,在典型工况下,SRS方法能够更好地捕捉到流动信息,具有较高的原始特性预测精度。尤其是SBES (Stress-blended eddy simulation)方法中的动态混合模型(Dynamic hybrid RANS-LES,DHRL)得到原始特性预测结果的最大误差仅为3.24%。此外提出了变黏度SRS热流场计算方法,发现在考虑工作介质黏度随温度变化后,相对恒黏度数值模拟结果,原始特性预测精度有明显改善,其最大误差小于3%。尺度解析模拟方法在液力变矩器性能预测方面的应用,有助于实现产品的低成本与短周期设计。

关键词: 尺度解析数值模拟, 内流场, 液力变矩器, 原始特性预测

Abstract: For the current situation that RANS method has always played a dominant role in numerical simulation of hydrodynamic torque converter, but it has low prediction accuracy and lacks accurate description of flow field structures, the SRS(scale-resolving simulation) method is employed to simulate the performance and internal flow field. The results show that SRS method can accurately capture the flow information and efficiently predict the original characteristics with high accuracy in the typical working conditions. Especially, the DHRL(dynamic Hybrid RANS-LES) model of SBES(stress-blended eddy simulation) method has shown the best prediction with the maximum error 3.24%. In addition, a method based on SRS model for calculating the thermal flow field with variable viscosity of transmission oil is also considered. It can be found that the performance prediction accuracy is obviously improved and the maximum error is less than 3% when the variable viscosity with temperature is considered. Applications of SRS analysis in the performance prediction of torque converter contributes to achieve low cost and short cycle design of the products.

Key words: hydrodynamic torque converter, internal flow field, performance prediction, scale-resolving simulation

中图分类号: