Rail waveform matching method based on dynamic time warping#br#

PDF(834 KB)

Journal of Vibration and Shock ›› 2018, Vol. 37 ›› Issue (11) : 246-251.

ZHU Hong-Tao，LI Shan，XIAO Yong，WEI Hui

Author information +

History +

Abstract

Rail flatness measuring instrument is a very important detecting tool in rail shortwave detection items. Due to its limited measurement range, the rail waveform to be measured needs the matching treatment to analyze the rail shortwave change law. Firstly, the coordinate transformation is done for the measured data to be unified in the same coordinate system. The traditional matching algorithm could be used only for the rigid transform and matching of curves in the same coordinate system, and it could not solve dynamic matching of curves in different coordinate systems. Here, a dynamic time warping (DTW) algorithm was proposed by bending and deforming the time axis to find the shortest bending path and the corresponding relation between point and point of curves to be matched. Whether the matching was good or bad was judged through calculating the shortest bending path. The rail flatness measuring instrument was used for the actual measurements of the standard rail. The traditional matching algorithm based on areas was compared with DTW. The test results showed that DTW algorithm can be used to realize the rail shortwave dynamic matching; its matching precision increases by 10 times compared with that of the traditional matching algorithm; it has a smaller matching error and a better stability.

Key words

rail shortwave detection / coordinate transformation / dynamic time warping / waveform matching

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

ZHU Hong-Tao，LI Shan，XIAO Yong，WEI Hui. Rail waveform matching method based on dynamic time warping#br#[J]. Journal of Vibration and Shock, 2018, 37(11): 246-251

References

[1] 程小红. 钢轨焊接接头平直度检测仪研制[D]. 西南交通大学,2010.
CHENG Xiao-hong.Research and Production of Flatness Gauge of Rail Wedding Joints[D].Southwest Jiaotong University,2010.
[2] 闫子权,谷爱军,刘维宁,V.L.Markine,梁青槐.
   Effects of wheelset vibration on initiation and evolution of rail short-pitch corrugation[J]. Journal of Central South University,2012,09:2681-2688.
[3] Jelali M.Performance assessment of control systems in rolling mills-application to strip thick thickness and flatness control. Jour-nal of Process Control . 2007
[4] Kong W,Kimia B B. On solving 2D and 3D puzzles using curve matching//Proceedings of the  CVPR.Hawaii, USA ,2001:583-590
[5] Ucoluk G, TorosluI H. Automatic reconstruction of broken 3-D surface objects. Computers and Graphics, 1999, 23(4):573-582
[6] 余先川,吕中华,胡丹.遥感图像配准技术综述[J].  光学精密工程,2013,11:2960-2972.
YU Chuan-xian,LV Zhong-hua,HU Dan.Reviwe of remote sensing image registration techniques[J].Optics and Precision Engineering,2013,11:2960-2972.
[7] 牛小兵,林玉池,赵美蓉等.基于特征的二维图像匹配法测量几何量[J].天津大学学报,2001,03:396-339.
NIU Xiao-bing,LIN Yu-chi,ZHAO Mei-rong etc. Measurement of geometric parameters based on feature based image mosaic method[J].Journal of TianJin University,2001,03:396-399.
[8] 肖辉,胡运发. 基于分段时间弯曲距离的时间序列挖掘[J]. 计算机研究与发展,2005,01:72-78.
XIAO Hui,HU Yun-fa.Data Mining Based on Segmented Time Warping Distance in Time Series Database[J].Journal of Computer Research and Development,2005,01:72-78.
[9]李正欣,张凤鸣,李克武等. 一种支持DTW距离的多元时间序列索引结构[J]. 软件学报,2014,03:560-575.
LI Zheng-xin,ZHANG Feng-ming,LI Ke-wu etc.Index Structure for Multivariate Time Series under DTW Distance Metric[J].Journal of Software,2014,03:560- 575.
[10] 邹朋成,王建东,杨国庆等. 辅助信息自动生成的时间序列距离度量学习[J]. 软件学报,2013,(11):2642-2655.
ZOU Peng-cheng,WANG Jian-dong,YANG Guo-qing etc. Distance Metric Learning Based on Side Information Autogeneration for Time Series[J].Journal of Software,2013,(11):2642-2655.
[11]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 of Similarity Search Based on Dynamic Time Warping in Time Series Data Mining[D].Harbin industrial university,2013
[12]邹朋成,王建东,杨国庆等. 辅助信息自动生成的时间序列距离度量学习[J]. 软件学报,2013,11:2642-2655.
ZOU Peng-cheng,WANG Jian-dong,YANG Guo-qing etc.Distance Metric Learning Based on Side Information Autogeneration for Time Series[J].Journal of software,2013,11:2642-2655.
[13]Prekopcsak Z,Lemire D. Time series classification by class-specific Mahalanobis distance measures. Advances in Data Analysis and Classification, 2012,6(3):185-200.
[14] Arash Gharehbaghi, Per Ask,Ankica Babic. A pattern recognition framework for detecting dynamic changes on cyclic time series[J]. Pattern Recognition, 2015, 48:696-708.
[15] 杜永峰,李万润,李慧等. 基于时间序列分析的结构损伤识别[J].振动与冲击,2012,(12):108-111.
DU Yong-feng,LI Wan-run,LI Hui etc. Structural damage identification based on time series analysis[J]. Journal of Vibration and Shock,2012,(12):108-111.
[16]严大勤,孙鑫. 一种基于区域匹配的图像匹配算法[J]. 仪器仪表学报,2006,S1:749-750+755.
YAN Da-qin,SUN Xin.Image mosaics algorithm based on area matching[J].Chinese Journal of Scientific Instrument,,2006,S1:749-750+755.
[17]顾费勇. 基于图像的自适应图像匹配算法研究[D].浙江大学,2008.
GU Fei-yong.Research on adaptive image mosaic algorithm based on image[D].Zhejiang University,2008.
[18]蒋瑜,陶俊勇,王得志等. 一种新的非高斯随机振动数值模拟方法[J]. 振动与冲击,2012,(19) :169-173
JIANG Yu,TAO Jun-yong,WANG Zhi-de etc. A novel approach for numerical simulation of a non-Gaussian random vibration[J].Journal of Vibration and Shock,2012,(19):169-173