车轮几何参数检测及误差分析

发布时间：2018-08-16 17:52

【摘要】：伴随着中国经济和科技实力的全面快速发展,国民经济和人民出行对交通运输的需求日益增长,轨道交通作为交通运输技术的重要一个分支,具备速度快、运量大、安全舒适、节能减排、效率高等无法媲美的优点。近十年来,轨道交通在国内外的发展突飞猛进,包括高速铁路、城际铁路、城市地铁。轨道交通建设的里程数增加速度也十分迅猛,运输量也快速上升。轨道交通发展在为国内经济发展、人们出行便利服务的同时,轨道交通的运输特点也发生了变化:运输任务加大、行车密度提升、运行速度提高,也正是因为这些变化,轮轨之间的相互作用也随之增大,加速了车轮轮对踏面的磨损,车轮几何参数的变化也随之加速。车轮几何参数的变化程度很大程度上决定了轮对的使用寿命,是线路大、中、小维修工作的重要部分,相关轨道公司的工务部门需要定期对车轮几何参数进行测量,掌握车轮几何参数变化情况,实现对其合理预测,进而相应地制定合理并具有针对性的维护计划。所以,对车轮几何参数实现高效准确的在线动态跟踪检测是一项十分必要的任务。本学位论文主要研究工作和创新成果,概括起来主要包括以下几个方面:1:介绍了车轮几何参数检测的主要内容,提出一种在线式动态跟踪检测方法,介绍测量的技术原理,基于激光三角法和图像处理技术设计的传感器,实现了对车轮轮对踏面轮廓非接触式高精度地检测,结合无线射频识别技术,提出了基于该技术的车轮几何参数动态在线跟踪检测方案,阐述了各个车轮几何参数的计算算法。2:车轮踏面轮廓图像处理,分析了车轮轮廓激光光带图像噪声来源和分类,利用空域滤波和频域滤波技术对车轮轮廓图像进行处理降噪,对处理后的图片进行比较分析,选择合适的降噪处理算法。车轮踏面轮廓检测的重要一个环节是对激光投射在轮对踏面形成的光带中心点提取,本文在对各种传统方法进行分析研究后,采用灰度重心法提取激光光带中心点。在此基础上,研究分析二维图像坐标系、传感器坐标系和世界坐标系三者之间的关系,得出了激光轮廓传感器的测量坐标系,根据轮廓上各点在测量坐标系坐标换算得到激光发射点的距离。3:在图像处理技术和激光轮廓传感器的基础上实现车轮几何参数在线动态跟踪检测后,针对列车在线动态行驶,钢轨竖向振动位移情况,分析了其对车轮几何参数中车轮滚动圆直径检测可能带来的误差进行了分析,建立有限元模型,分析轨道振动,结合车轮滚动圆动态在线测量原理,在ANSYS下进行模拟仿真计算,得出钢轨竖向振动对其测量可能带来的影响。4:根据提出的车轮几何参数在线动态跟踪检测方法,现场布置传感器,进行现象实验,并得出结果。
[Abstract]:With the rapid development of China's economy and science and technology, the demand of national economy and people for transportation is increasing day by day. As an important branch of transportation technology, rail transit has the advantages of fast speed, large capacity, safety and comfort. Energy saving and emission reduction, high efficiency and other unparalleled advantages. In the past ten years, rail transit has developed rapidly at home and abroad, including high speed railway, intercity railway and city subway. The mileage of rail transit construction is also increasing rapidly, and the volume of transportation is also rising rapidly. With the development of rail transit for the development of domestic economy and the convenient service for people to travel, the transport characteristics of rail transit have also changed: the transportation task has been increased, the traffic density has been increased, and the speed of operation has been increased, which is precisely because of these changes. The interaction between wheel and rail increases, which accelerates the wear of wheel tread and the change of wheel geometric parameters. The variation of wheel geometric parameters determines the service life of wheelset to a great extent, and is an important part of the maintenance work of large, medium and small lines. The public works departments of relevant rail companies need to measure the geometric parameters of wheels regularly. According to the change of wheel geometry parameters, the reasonable forecast is realized, and the reasonable and targeted maintenance plan is worked out accordingly. Therefore, it is a very necessary task to realize efficient and accurate on-line dynamic tracking detection of wheel geometry parameters. In this dissertation, the main research work and innovative achievements are summarized as follows: the main contents of wheel geometric parameter detection are introduced, a on-line dynamic tracking detection method is proposed, and the technical principle of measurement is introduced. Based on the sensor designed by laser triangulation and image processing technology, the non-contact high-precision detection of wheel tread profile is realized, and the wireless radio frequency identification (RFID) technology is combined. This paper presents a dynamic on-line tracking detection scheme for wheel geometric parameters based on this technique. The calculation algorithm of each wheel geometry parameter is described. The image processing of wheel tread profile is described. The noise source and classification of wheel profile laser belt image are analyzed. Spatial filtering and frequency domain filtering are used to process the noise of wheel contour image. The image after processing is compared and analyzed, and the appropriate denoising algorithm is selected. One of the most important steps in wheel tread profile detection is to extract the center of the laser beam which is formed on the wheel tread. After analyzing and studying the traditional methods, the gray gravity method is used to extract the center of the laser beam. On this basis, the relationship among two-dimensional image coordinate system, sensor coordinate system and world coordinate system is analyzed, and the measuring coordinate system of laser contour sensor is obtained. The distance of laser emission point is obtained according to the coordinate conversion of each point on the contour in the measuring coordinate system. On the basis of image processing technology and laser contour sensor, the on-line dynamic tracking detection of wheel geometric parameters is realized, and the train is running dynamically on line. The vertical vibration displacement of rail is analyzed. The error caused by wheel rolling circle diameter detection in wheel geometry parameters is analyzed. The finite element model is established to analyze the track vibration, and the dynamic on-line measurement principle of wheel rolling circle is combined. The influence of vertical vibration of rail on the measurement of rail is obtained by simulation calculation under ANSYS. 4. According to the proposed on-line dynamic tracking detection method for wheel geometry parameters, sensors are arranged on the spot, and the phenomena are tested, and the results are obtained.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U270.7;TP391.41

【参考文献】