基于动态时间规整的钢轨轨廓匹配方法

PDF(2158 KB)

振动与冲击 ›› 2019, Vol. 38 ›› Issue (20) : 210-215.

论文

基于动态时间规整的钢轨轨廓匹配方法

陆英杰，朱洪涛

作者信息 +

A rail profile matching method based on dynamic time warping

LU Yingjie, ZHU Hongtao

Author information +

文章历史 +

摘要

钢轨轨廓的精准匹配是研究其几何形态变化的重要前提。由于轨检小车行进时产生随机振动，引起传感器振动，导致采集数据失真。在测量过程中，因无法保证传感器绝对对称，故每个传感器的振动存在一些差异性。传统的特征点算法只能对振动引起的噪声进行刚性补偿，且有时特征点提取不到。基于此，提出动态时间规整算法，从初步匹配的数据中，选取标准轨廓和测量轨廓轨腰处数据作为两个时间序列，建立距离矩阵，通过对时间轴的扭曲变形，利用递归思想，计算累积代价矩阵，得到最优规整路径，从而找到两个时间序列各点的对应关系，实现轨廓的精准匹配。实验结果表明，相比特征点算法，动态时间规整算法的匹配精度提高了近7倍，匹配效果较好，稳定性很高。

Abstract

Precise matching of rail profiles is an important prerequisite for studying changes in their geometric shapes.Due to random vibration is generated when the track inspection trolley travels, the vibration of the sensor is caused, the acquired data are distorted.In the measurement process, there is some difference in the vibration of each sensor because the absolute symmetry of the sensor cannot be guaranteed.The traditional feature point algorithm can only rigidly compensate for vibration-induced noise, and sometimes feature points cannot be extracted.Based on this, a dynamic time warping algorithm was proposed.From the preliminary matching data, the data of the standard rail profile and the lumbar profile of the measurement rail were selected as two time series, and the distance matrix was established.The recursive idea was used to calculate the cumulative cost matrix with twisting the time axis.The optimal regulatory path was obtained.The correspondence was found between each point in the two time series so as to achieve the exact match of the track profile.Experimental results show that compared with the feature point algorithm, the matching accuracy of the dynamic time warping algorithm is improved by nearly 7 times, the matching effect is good, and the stability is high.

导出引用

陆英杰，朱洪涛. 基于动态时间规整的钢轨轨廓匹配方法[J]. 振动与冲击, 2019, 38(20): 210-215

LU Yingjie, ZHU Hongtao . A rail profile matching method based on dynamic time warping[J]. Journal of Vibration and Shock, 2019, 38(20): 210-215

参考文献

[1]SANTUR Y, Karakose M, Akin E. Big data framework for rail inspection[C]// International Artificial Intelligence and Data Processing Symposium. 2017:1-4.
[2]郭雁一夫. 钢轨轮廓检测系统的设计与实现[D].湖南大学,2016.
Guo Yanyifu. Design and implementation of rail profile detection system [D]. Hunan University, 2016.
[3]JIN W, Zhan X, Jiang B. Non-contact Rail-wear Inspecting System Based on Image Understanding[C]// International Conference on Mechatronics and Automation. IEEE, 2007:3854-3858.
[4] SANTUR Y, Karakose M, Akin E. A new rail inspection method based on deep learning using laser cameras[C]// International Artificial Intelligence and Data Processing Symposium. 2017:1-6.
[5]WEN Chen,Hai Feng Li,Yu De Xu. Method on Whole Rail Profile Auto-Matching and its Application in Urban Rail Transit[J]. Applied Mechanics and Materials,2012,1976(209).
[6]MOLLEDA J, Usamentiaga R, ýlvaro F. Millara, et al. A profile measurement system for rail manufacturing using multiple laser range finders[C]// Industry Applications Society Meeting. IEEE, 2015:1-8.
[7]康高强,李春茂,秦莉娟,马俊.一种钢轨动态轮廓数据校准方法研究[J].传感技术学报,2015,28(02):221-226.
KANG Gaoqiang,LI Chunmao,QIN Lijuan,Ma Jun.Research on A Method for Calibrating Dynamic Rail Profile Data[J].Chinese Journal of Transduction Technology,2015,28(02):221-226.
[8] 占栋,于龙,肖建,卢明舫.钢轨轮廓全断面检测中轨廓动态匹配方法研究[J].铁道学报,2015,37(05):71-77.
ZHAN Dong,Yu Long,XIAO Jian,LU Mingqi.Study on High—accuracy Vision Measurement Approach for Dynamic Inspection of Full Cross-sectional Rail Profile[J].Journal of the China Railway Society,2015,37(05):71-77.
[9] 谭周文,郭雁一夫.激光三角法钢轨磨耗检测系统研究[J].传感器与微系统,2016,35(05):20-24.
TAN Zhouwen,GUO Yanyifu.Study on Rail Wear Detection System of laser triangulation method[J].Transducer and Microsystem Technologies,2016,35(05):20-24.
[10]严隽耄．车辆工程［Ｍ］．北京：中国铁道出版社，２００９．
YAN Yanmao. Vehicle Engineering [M]. Beijing: China Railway Publishing House, 2009.
[11]占栋,于龙,邱存勇,肖建,陈唐龙.钢轨轮廓测量中的车体振动补偿问题研究[J].仪器仪表学报,2013,34(07):186-194.
ZHAN Dong,YU Long,QIU Cunyong,XIAO Jian,CHEN Tanglong.Research on Vehicle Body Vibration Compensation in Rail Contour Measurement[J].Chinese Journal of Scientific Instrument,2013,34(07):186-194.
[12]王敏,童水光,陈玉辉,从飞云.一种基于Hough变换的快速圆检测算法[J].机械工程与自动化,2018(01):152-154.
WANG Min,TONG Shuiguang,CHEN Yuhui,CONG Feiyun.A Quick Circle Detection Algorithm Based on Hough Transform[J].Mechanical Engineering & Automation,2018(01):152-154.
[13]杨洁, 康宁. 动态时间规整DTW算法的研究[J]. 科技与创新, 2016(4):11-12.
YANG Jie, KANG Ning. Research on Dynamic Time Warping DTW Algorithm[J]. Science and Technology & Innovation, 2016(4):11-12.
[14]杨靖. 基于动态时间弯曲的时间序列相似性搜索技术的研究[D].哈尔滨工业大学,2013.
YANG Jing. Research on time series similarity search based on dynamic time warping[D]. Harbin Institute of Technology, 2013.
[15]鞠标,朱洪涛,徐鞠萍,王志勇,孙政.基于轨检车振动问题的钢轨轮廓匹配方法研究[J].振动与冲击,2017,36(03):65-69.
JU Biao,ZHU Hongtao,XU Yuping,WANG Zhiyong,SUN Zheng.Rail Contour Matching Method Based on Random Vibrations of A Rail Inspection Vehicle [J].Journal of Vibration and Shock,2017,36(03):65-69.
[16]李海林,梁叶,王少春.时间序列数据挖掘中的动态时间弯曲研究综述[J/OL].控制与决策:1-10[2018-03-26].https://doi.org/10.13195/j.kzyjc.2017.1037.
LI Hailin, LIANG Ye, WANG Shaochun. A Review on Dynamic Time Wraping in Time Series Data Mining [J/OL]. Control and Decision: 1-10[2018-03-26]. https://doi.org/10.13195/j .kzyjc.2017.1037.
[17]李正欣,张凤鸣,李克武,张晓丰.一种支持DTW距离的多元时间序列索引结构[J].软件学报,2014,25(03):560-575.
LI Zhengxin,ZHANG Fengming,LI Kewu, ZHANGXiaofeng. Index Structure for Multivariate Time Series under DTW Distance Metric[J].Journal of Software,2014,25(03):560-575.