基于结构光视觉传感器的轮对踏面几何参数在线测量系统研究

发布时间：2018-03-09 21:34

本文选题：在线检测　切入点：误差分析　出处：《北京交通大学》2017年硕士论文　论文类型：学位论文

【摘要】：进入21世纪,中国铁路建设各项工作取得显著成绩,中国铁路进入了新的发展时期。火车车轮是铁路机车车辆的主要运动部件,形状较为复杂,现有的常规无损检测方法只能对其中某些部分进行检测,对于车轮踏面表面和近表面区域中参数的完整检测尚未有一种较为理想的方法,因此,随着列车速度的不断提升,研究轮对踏面几何参数检测技术与系统具有重要的现实意义。本文基于双二维结构光视觉传感器设计并构建了轮对踏面几何参数检测系统,使用该检测系统采集轮对踏面轮廓数据,并对数据进行综合分析处理,获得踏面几何参数,同时对影响测量的误差因素进行讨论,给出相应的补偿措施。结果表明,基于双二维结构光视觉传感器的踏面轮廓几何参数测量系统方案可行性,测量结果满足轮对检测的各项指标要求。本文的工作内容如下:(1)搜集查阅了国内外列车轮对检测系统的进展和成果,掌握轮对踏面外形测量技术的研究现状。(2)设计了基于双二维结构光传感器的轮对踏面几何参数检测方案,详细介绍踏面参数测量原理,进行了视觉传感器选型以及参数配置。(3)搭建了基于双二维结构光传感器的轮对踏面几何参数检测系统,实现了对轮对主要几何参数的测量,其中轮缘几何参数测量精度达到±0.2mm,轮辋宽精度达到±0.4mm。(4)对引起误差的来源及其对测量精度的影响程度进行了分析。(5)将实验室结论应用于地铁检测现场,验证了系统在线测量的可行性。
[Abstract]:In 21th century, China's railway construction has made remarkable achievements, and the Chinese railway has entered a new period of development. The train wheels are the main moving parts of railway locomotives and rolling stock, and the shapes are quite complex. The existing conventional nondestructive testing methods can only detect some parts of them, but there is no ideal method for the complete detection of the parameters in the surface and near surface of the wheel tread, so with the increasing of the train speed, there is no ideal method for the complete detection of the parameters of the wheel tread surface and the near surface area. It is of great practical significance to study the technology and system of wheel tread geometric parameter detection. In this paper, a wheel tread geometric parameter detection system is designed and constructed based on double two-dimensional structured light vision sensor. The measuring system is used to collect the profile data of the wheel pair tread, and the data are analyzed and processed synthetically, and the geometric parameters of the tread are obtained. At the same time, the error factors affecting the measurement are discussed, and the corresponding compensation measures are given. The scheme of measuring geometric parameters of tread profile based on dual two-dimensional structured light vision sensor is feasible. The measurement results meet the requirements of the wheelset test. The work of this paper is as follows: 1) the progress and achievements of train wheelset detection system at home and abroad have been collected and consulted. Master the research status of wheel tread profile measurement technology. (2) Design the measuring scheme of wheel tread geometry parameter based on double two-dimensional structure light sensor, and introduce the principle of tread surface parameter measurement in detail. In this paper, the selection of vision sensor and the configuration of parameters are carried out. The measuring system of wheel tread geometric parameters based on double two-dimensional structure light sensor is set up, and the main geometric parameters of wheelset are measured. The accuracy of wheel rim geometric parameter measurement is 卤0.2mm, the rim width accuracy is 卤0.4mm. 4) the source of the error and the influence degree on the measurement accuracy are analyzed. The laboratory results are applied to the subway detection site. The feasibility of on-line measurement is verified.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U279.32;TP212.9

【参考文献】