现代制造工程 ›› 2026, Vol. 544 ›› Issue (1): 101-107.doi: 10.16731/j.cnki.1671-3133.2026.01.012

• CAD/CAE/CAPP/CAM • 上一篇    下一篇

重型液力自动变速箱智能换挡策略的联合仿真研究*

何俊瑶1, 吴怀超1, 聂龙2, 刘波3, 谢嘉3, 龙运祥3, 赵丽梅1   

  1. 1 贵州大学机械工程学院,贵阳 550025;
    2 贵州理工学院机械工程学院,贵阳 550003;
    3 贵州航天凯星智能传动有限公司,遵义 563114
  • 收稿日期:2025-02-06 出版日期:2026-01-18 发布日期:2026-03-17
  • 作者简介:何俊瑶,硕士研究生,主要研究方向为重型液力自动变速器的换挡策略。吴怀超,博士,教授,博士生导师,主要研究方向为流体传动与控制技术等。E-mail:1255246283@qq.com
  • 基金资助:
    *贵州省全省重点实验室项目(黔科合平台ZSYS〔2025〕002);贵州省高等学校重点实验室项目(黔教技[2023]009号);贵州省科技计划项目(黔科合成果[2023]一般768);贵州省基础研究计划项目(黔科合基础-ZK[2023]一般133)

Research on joint simulation technology of intelligent shifting strategy of heavy-duty AT

HE Junyao1, WU Huaichao1, NIE Long2, LIU Bo3, XIE Jia3, LONG Yunxiang3, ZHAO Limei1   

  1. 1 School of Mechanical Engineering, Guizhou University, Guiyang 550025, China;
    2 School of Mechanical Engineering, Guizhou Institute of Technology, Guiyang 550003, China;
    3 Guizhou Winstar Hydraulic Transmission Machinery Co., Ltd., Zunyi 563114, China
  • Received:2025-02-06 Online:2026-01-18 Published:2026-03-17

摘要: 在重型液力自动变速器的研究中,现有的三参数自动换挡策略难以满足实际车辆在复杂工况下的换挡需求,尤其是在载重和坡度变化时,其自适应性不足。基于此,提出了一种利用Simulink软件搭建 AMEsim整车模型中的电子控制单元(ECU)的联合仿真技术。首先,将BP的工作原理应用于某型自动变速箱的换挡过程,并在MATLAB/Simulink中搭建整车模型,从该模型的数据中获取数据样本用于训练BP;然后,在AMEsim中搭建整车模型,其中变速器控制单元(TCU)用神经网络拟合工具箱训练出来的Simulink模型替代,实现联合仿真;最后,通过对比三参数换挡和智能换挡策略在动力性和经济性方面的表现,结果表明智能换挡比三参数自动换挡具有更高的精度和自适应能力。为自动变速器的智能控制提供了理论基础和技术支持。

关键词: 重型液力自动变速器, MATLAB, AMEsim, 联合仿真

Abstract: In the study of heavy-duty hydraulic automatic transmissions,the existing three-parameter automatic shifting strategy fails to meet the shifting requirements of real-world vehicles,and is unable to dynamically adjust based on load and gradient. Its adaptability,therefore,requires improvement. Based on this, a joint simulation technology for the ECU in the AMEsim vehicle model was proposed using Simulink software. First,the working principle of BP was applied to the shifting process of a specific automatic transmission,and the vehicle model was constructed in MATLAB/Simulink software. Data samples were extracted from the model for BP training,after which the vehicle model was constructed in AMEsim. The TCU was replaced with a Simulink model trained using the neural network fitting toolbox to enable joint simulation. Finally,the dynamic performance and economic efficiency of the three-parameter and intelligent shifting strategies were compared. The results show that the intelligent shifting strategy offers greater accuracy and adaptability than the three-parameter automatic shifting, providing a theoretical foundation and technical support for the intelligent control of automatic transmissions.

Key words: heavy-duty hydraulic automatic transmission, MATLAB, AMEsim, joint simulation

中图分类号: 

版权所有 © 《现代制造工程》编辑部 
地址:北京市东城区东四块玉南街28号 邮编:100061 电话:010-67126028 电子信箱:2645173083@qq.com
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持:support@magtech.com.cn