减轻运行过程中振动对车载钢轨轮廓检测系统的不利影响,获取实时准确的轮廓检测数据是钢轨维护工作面临的重要问题.在对比激光视像技术和激光位移技术轮廓检测原理异同的基础上,对更适合高速铁路检测精度需求的激光位移技术进行了系统研究,并利用二维激光位移传感器搭建了钢轨轮廓动态测量平台;针对位移技术在动态测量时一直未能有效解决的变形轮廓校正问题,指出传统视像技术的解决方法不能被机械地照用,并提出一种基于轨颚点与轨腰特征区域双重匹配的轮廓仿射变换参数估计方法,然后利用粒子群算法对变换参数进行迭代优化,实现钢轨变形轮廓校准,最后通过模拟振动实验验证了本方法的有效性.实验结果表明,本方法能较为准确地消除振动对测量数据的影响,为激光位移技术在钢轨轮廓高精度动态检测中的推广应用提供了新的思路和技术参考.%Through alleviating the adverse influence of vibration on on-board rail profile detection system to col-lect the accurate and real-time profile parameters is a key challenge to rail maintenance.On the basis of the comparison with the laser visual technique and the laser displacement technique for rail profile detection,a sys-tematic research on the latter was carried out,and a corresponding measurement platform was built with the 2D laser displacement sensors.Because the traditional solutions based on the laser visual technique cannot be copied mechanically for solving the hard problem on distorted profile calibration faced by the laser displacement technique,a novel method for estimating the affine transformation parameters of rail profile was put forward in this paper based on the double match including the rail jaw point and rail waist area.The particle swarm algo-rithm was employed iteratively for optimizing the transformation parameters,and the profile calibration was completed finally.The validity of this method was verified by the experiment with vibration.The results show the influence of vibration on the measurement data can be eliminated effectively by our proposed method.It provides a new idea and technical reference for the popularization and application of the laser displacement tech-nique in the high-precision dynamic rail profile measurement.
展开▼
