• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2026, Vol. 62 ›› Issue (16): 383-391.doi: 10.3901/JME.260718

• 特邀专辑:轨道运载装备动态设计理论与运维技术 • 上一篇    

扫码分享

基于机器视觉的铁道车辆二系横向止挡工作间隙检测方法

周义昌1, 彭晨1, 王开云2, 陈吉永3, 凌亮2, 邱知恒1, 任尊松1   

  1. 1. 北京交通大学机械与电子控制工程学院 北京 100044;
    2. 西南交通大学轨道交通运载系统全国重点实验室 成都 610031;
    3. 中车大同电力机车有限公司 大同 037038
  • 收稿日期:2025-08-13 修回日期:2026-03-07 发布日期:2026-10-14
  • 作者简介:周义昌(通信作者),男,1993年出生,博士,讲师。主要研究方向为轨道车辆动力学和智能运维。E-mail:yczhou@bjtu.edu.cn
  • 基金资助:
    中央高校基本科研业务费专项(2025JBMC020)、国家自然科学基金(52405087,52388102)和国家重点研发计划(2025YFF0524300)资助项目。

Machine Vision-based Method for Detecting the Working Clearance of Lateral Bumpstops of Railway Vehicles

ZHOU Yichang1, PENG Chen1, WANG Kaiyun2, CHEN Jiyong3, LING Liang2, QIU Zhiheng1, REN Zunsong1   

  1. 1. School of Mechanical, Electronic and Control Engineering, Beijing Jiaotong University, Beijing 100044;
    2. State Key Laboratory of Rail Transit Vehicle System, Southwest Jiaotong University, Chengdu 610031;
    3. CRRC Datong Electric Locomotive Co., Ltd., Datong 037038
  • Received:2025-08-13 Revised:2026-03-07 Published:2026-10-14

摘要: 二系横向止挡是限制铁道车辆车体和转向架横向相对位移的关键部件,因其允许的安装空间有限,且止挡工作间隙变化频繁,缺少一种能长期监测其工作状态的有效手段。以某型重载机车为研究对象,提出一种基于机器视觉的铁道车辆二系横向止挡工作间隙的检测方法,包括图像采集、图像增强、边缘检测、概率霍夫变换等主要流程,并研究边缘检测中高低阈值和霍夫变换中累加器阈值等重要参数对止挡工作间隙检测结果的影响规律,获得图像处理的最优参数组合。结果表明,在光线条件良好的情况下,所提出的基于机器视觉的检测方法能有效监测二系横向止挡的工作状态,与拉线位移传感器的测量结果相比,平均偏差小于10%,为止挡工作间隙的长期在线监测提供一种新的技术思路。

关键词: 铁道车辆, 横向止挡, 机器视觉, 工作间隙, 边缘检测

Abstract: The secondary lateral bumpstop is a key component that limits the lateral relative displacement between the car-body and the bogie in railway vehicles. However, due to restricted installation space and frequent variations in working clearance, no effective method currently exists for monitoring its operational status. A machine vision-based approach for detecting the working clearance of the secondary lateral stop is introduced when a locomotive is taken as the interested object. The method involves image acquisition, image enhancement, edge detection, and probabilistic Hough transform, and the effects of key parameters in edge detection and Hough transform, such as high, low and accumulator thresholds, on detection accuracy are systematically analyzed. Experimental results demonstrate that the proposed method can effectively monitor the operational status of the secondary lateral stop, with detection errors within 10% compared to measurements obtained from a cable displacement sensor. It provides a novel approach for the long-term online monitoring of the working clearance of lateral bumpstops.

Key words: railway vehicles, lateral bumpstop, machine vision, working clearance, edge detection

中图分类号: