汽车工程 ›› 2025, Vol. 47 ›› Issue (10): 2016-2026.doi: 10.19562/j.chinasae.qcgc.2025.10.017

• • 上一篇    

智能轮胎磨损敏感区域信号特征及估算方法

王国林1,2,王鑫1,荆哲铖1(),李相良1,张宇1   

  1. 1.江苏大学汽车与交通工程学院,镇江 212013
    2.江苏大学京江学院,镇江 212013
  • 收稿日期:2024-12-25 修回日期:2025-03-04 出版日期:2025-10-25 发布日期:2025-10-20
  • 通讯作者: 荆哲铖 E-mail:jingzc@ujs.edu.cn
  • 基金资助:
    国家自然科学基金(52272366)

Sensitive Area of Tire Wear Signal Characteristics and Wear Estimation Method for Intelligent Tire

Guolin Wang1,2,Xin Wang1,Zhecheng Jin1(),Xiangliang Li1,Yu Zhang1   

  1. 1.School of Automotive and Traffic Engineering,Jiangsu University,Zhenjiang 212013
    2.Jingjiang College,Jiangsu University,Zhenjiang 212013
  • Received:2024-12-25 Revised:2025-03-04 Online:2025-10-25 Published:2025-10-20
  • Contact: Zhecheng Jin E-mail:jingzc@ujs.edu.cn

摘要:

轮胎磨损不仅关乎车辆行驶安全,还对轮胎物理模型参数优化具有重要影响。本文提出了一种可应用于应变型智能轮胎的轮胎磨损状态估算方法。首先,借助有限元技术获取运动轮胎内衬层周向应变,并分析磨损对其影响机理,提出4个与轮胎磨损密切相关的信号特征指标。接着,基于全局灵敏度理论,探究了这些磨损特征指标对轮胎使用状态(磨损、胎压、车速和载荷)的敏感程度及内衬层敏感区域。结果表明,轮胎内衬层中心点对周向应变1阶导数的磨损特征指标最为敏感,而中心点两侧17~27 mm处对周向应变的磨损特征指标最为敏感,此结论可用于指导传感器安装。在此基础上利用高斯过程回归建立轮胎磨损状态估算模型,考虑轮胎使用状态开发的模型估算结果的平均RMSE为0.166 mm,此方法不仅确保了估算精度,还充分利用了车辆行驶过程中的既有数据资源,实现了对轮胎磨损状态的有效监控与管理。

关键词: 智能轮胎, 轮胎磨损, 全局灵敏度分析, 轮胎应变, 高斯过程回归

Abstract:

Tire wear not only affects vehicle driving safety, but also has an important impact on the optimization of tire physical model parameters. In this paper, a tire wear state estimation method that can be applied to strain type intelligent tires is proposed. Firstly, the circumferential strain of the inner liner of the moving tire is obtained by using finite element technology and the impact mechanism of wear on it is analyzed. Four feature indicators closely related to tire wear are proposed. Then, based on the global sensitivity indicator theory, the sensitivity of these wear features to tire using conditions (wear, tire pressure, vehicle speed, and load) and the inner liner sensitive area are explored. The results show that the first derivative of the circumferential strain at the center point of the tire inner liner is the most sensitive to wear features, while the circumferential strain at 17-27 mm on either side of the center point is the most sensitive to wear features, which can be used to guide the sensor installation position. Finally, the Gaussian process regression is used to develop the wear state estimation model, and the average RMSE of the model estimation results considering the tire use conditions is only 0.166 mm. This method not only ensures the estimation accuracy, but also makes full use of the established data resources during the vehicle driving process, ensuring effective monitoring and management of tire wear state.

Key words: intelligent tire, tire wear, global sensitivity analysis, tire strain, Gaussian process regression