基于四点弦测法的轨面不平顺检测及复原方法  

Rail Surface Irregularities Detection and Restoration Based on Four-point Chord Reference Method

作　　者：毛晓君[1] 许玉德[1] 周宇[1] 

机构地区：[1]同济大学道路与交通工程教育部重点实验室,上海201804

出　　处：《华东交通大学学报》2013年第5期13-17,共5页Journal of East China Jiaotong University

基　　金：国家自然基金项目(50908179);中央高校基本科研业务费专项资金资助项目

摘　　要：为了实现轨面不平顺的在线连续检测,并进一步提高弦测法的检测精度,提出了一种四点弦测法。通过理论推导其传递函数,并基于快速傅里叶(FFT)变换利用MATLAB建立了其自动复原算法。为更加合理地确定四点弦测法各参数,用20个随机产生的正弦波叠加模拟轨面不平顺并选用复原误差的最大值、平均值和标准差3个指标来进行复原精度比选。在最优参数的基础上,对上海地铁一段长30 m的实测数据进行复原,其结果与对应形式下三点弦测法的复原结果对比,结果表明四点弦测法的复原精度能控制在微米级,且相对于三点弦测法误差平均值和标准差均改善了50%以上,最大值改善了70%以上。In order to realize continuous detection of rail surface irregularities and improve detection precision, a new four-point approach based on chord reference method is presented in this paper. With the theoretical de-duction of its transfer function,the automatic restoration algorithm based on Fast Fourier Transform(FFT)has been realized by means of MATLAB. A simulation waveform consisting of 20 random sine waves has been used for scientific determination of the three separation distances. The maximum value,average value and stan-dard deviation of restoration error are therefore adopted to weigh their restoration precision. Finally,on the ba-sis of optimal parameters,a length of 30m real track irregularities from Shanghai metro is used for comparison. Results show that restoration precision of the new four-point method has been controlled at micron level. In the aspect of restoration error’s average value and standard deviation,the detection precision has been improved over 50%and the maximum value has been improved over 70%when compared to the traditional three-point method.

关 键 词：轨面不平顺 弦测法 复原算法 复原精度 

分 类 号：U25[交通运输工程—道路与铁道工程]