• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2025, Vol. 61 ›› Issue (18): 406-416.doi: 10.3901/JME.2025.18.406

• 交叉与前沿 • 上一篇    

扫码分享

液力变矩器涡识别方法和介质流动特性研究

郑秋霖1, 柴博森1,2,3, 耿冬妮1, 范文赫1, 李涛1, 冉子林4   

  1. 1. 吉林大学机械与航空航天工程学院 长春 130025;
    2. 吉林大学重庆研究院 重庆 401120;
    3. 吉林大学汽车底盘集成与仿生全国重点实验室 长春 130025;
    4. 重庆交通大学机电与车辆工程学院 重庆 400074
  • 收稿日期:2024-10-10 修回日期:2025-03-03 发布日期:2025-11-08
  • 作者简介:郑秋霖,男,1999年出生,博士研究生。主要研究方向为流体传动与控制。E-mail:zhengql22@mails.jlu.edu.cn;柴博森(通信作者),男,1984年出生,博士,教授,博士研究生导师。主要研究方向为流体传动与控制。E-mail:chaibs2012@jlu.edu.cn
  • 基金资助:
    国家自然科学基金(52075212)、重庆市自然科学基金(CSTB2023NSCQ-MSX0321)、吉林省教育厅科学研究(JJKH20250078KJ)、吉林大学汽车底盘集成与仿生全国重点实验室自由探索(ascl-zytsxm-202010)、吉林大学“大学生创新创业训练计划”(202410183135)和浙江省博士后科研择优(ZJ2024164)资助项目

Rheological Properties of Torque Converter Media and Applicability of Three-dimensional Vortex Identification Method

ZHENG Qiulin1, CHAI Bosen1,2,3, GENG Dongni1, FAN Wenhe1, LI Tao1, RAN Zilin4   

  1. 1. School of Mechanical and Aerospace Engineering, Jilin University, Changchun 130025;
    2. Chongqing Research Institute, Jilin University, Chongqing 401120;
    3. National Key Laboratory of Automotive Chassis Integration and Bionics, Jilin University, Changchun 130025;
    4. School of Electromechanical and Vehicle Engineering, Chongqing Jiaotong University, Chongqing 400074
  • Received:2024-10-10 Revised:2025-03-03 Published:2025-11-08

摘要: 液力变矩器内部非定常多尺度涡动力学特性支配并影响其流场结构演化过程。基于计算流体动力学理论,采用应力混合涡湍流模型(Stress-blended eddy simulation,SBES)数值模拟失速工况下液力变矩器油介质与水介质流场,对比分析不同介质下流场结构的相似性及差异性,综合评价以水介质开展流场演化规律科学研究的可行性,通过粒子图像测速(Particle image velocimetry,PIV)可视化试验验证水介质仿真结果的准确性。研究采用不同涡识别方法提取液力变矩器非定常多尺度涡系结构,从流道空间整体重构效果及叶片近壁面湍流局部结构辨识两个方面,分析评价不同涡识别方法。结果表明:① Q准则阈值选择盲目性大,涡系重构效果依赖经验;Ω方法涡系整体重构效果好,但是针对叶片近壁面区域多尺度涡系辨识能力薄弱,弱涡结构特征识别丢失;Ω-Liutex方法空间多尺度涡系重构效果最佳,强涡、弱涡结构细节特征提取能力强,阈值选择敏感性低,涡系重构效率高;② 油介质与水介质整体流场仿真结果具有高度相似性,但是因介质黏度影响,叶片近壁面局部流场特征存在差异性,油介质涡系重构更加饱满,叶轮之间流动连续性更强。研究结果可为液力变矩器流场演化规律解析及叶栅结构正向设计提供一定的技术指导。

关键词: 液力变矩器, 计算流体动力学, 粒子图像测速, 涡识别方法, 阈值选择

Abstract: The non-stationary multiscale eddy dynamics inside the torque converter governs and influences the evolution of its flow field structure. Based on the theory of computational fluid dynamics, the stress-blended eddy simulation(SBES) is used to numerically simulate the flow field of the torque converter in oil medium and water medium under the stall condition, to compare and analyse the similarity and difference of the flow field structure in different media, and to comprehensively evaluate the feasibility of carrying out a scientific study of the flow field evolution law in water medium. The accuracy of the simulation results in aqueous medium is verified by particle image velocimetry(PIV) visualisation test. Different vortex identification methods are used to extract the non-constant multi-scale vortex structure of the torque converter, and the different vortex identification methods are analysed and evaluated in terms of the overall reconstruction effect of the flow channel space and the identification of the turbulent local structure of the near-wall surface of the blade. The results show that:① Q criterion threshold selection blindness, vortex reconstruction effect depends on experience;Ω method vortex reconstruction effect is good, but for the blade near the wall region of the multi-scale vortex identification ability is weak, weak vortex structural features are lost;Ω-Liutex method spatial multi-scale vortex reconstruction effect is the best, strong vortex, weak vortex structural details feature extraction ability, threshold selection sensitivity is low, vortex reconstruction efficiency is high. ② The simulation results of the overall flow field of oil medium and water medium are highly similar, but due to the influence of medium viscosity, there are differences in the local flow field characteristics of the near-wall surface of the blade, and the vortex system reconstruction of the oil medium is more full, and the continuity of the flow between impellers is stronger. The results of the study can provide certain technical guidance for the analysis of the flow field evolution law of the torque converter and the forward design of the blade grid structure.

Key words: torque converter, computational fluid dynamics, particle image velocimetry, vortex identification method, threshold selection

中图分类号: