摘要
该文针对传统弦测方法有效检测波段较窄,不能通过单个系统对30mm^60m全波段弦测波形进行复原的缺陷,提出一种基于两点弦与三点偏弦相结合的钢轨波磨广域测量方法。该方法一方面使用最小二乘法来设计逆滤波器,只需计算传递函数的幅频特性,较传统的频率采样法省去了相频特性的计算;另一方面在三点偏弦法的架构下,通过3个检测点灵活组合,将两点弦长波测量与三点偏弦短波测量优势进行互补,实现了单个系统下的全波段弦测波形的准确复原与测量。为验证该方法的有效性,在室内轨检小车上利用1D激光位移传感器搭建了钢轨波磨测量平台,并进行了线路测试。实验结果表明,该方法能较为精确地对广域轨道波磨进行测量。
In view of the defect that the effective detection waveband of the traditional chord-based method is too narrow to be used for recovering the corrugation of the full waveband whose wavelength range varies from 30mm to 60m,a wide-area measurement method of rail corrugation based the combination of two-point and three-point chord offset measuring methods is proposed in this paper.The method based on LS to design the inverse filter.Compared with the traditional frequency sampling method,only the amplitude-frequency characteristic of the transfer function is calculated without calculating the phase-frequency characteristic.Under the architecture of three-point chord offset system,the advantages of two methods are complemented through the flexible combination of 3 detection positions.Then,by selecting the length of chord and division ratio properly,the corrugation of the full waveband is recovered and measured accurately based a single system.The effectiveness is verified via the experiment conducted on an indoor measurement platform built with 1D laser displacement sensors.The results compared with manual gauge show that the proposed method can measure the wide-area rail corrugation precisely.
作者
魏珲
刘宏立
马子骥
李艳福
WEI Hui;LIU Hongli;MA Ziji;LI Yanfu(College of Electrical and Information Engineering,Hunan University,Changsha 410082,China)
出处
《西北大学学报（自然科学版）》
CAS
CSCD
北大核心
2018年第2期199-208,共10页
Journal of Northwest University（Natural Science Edition）
基金
中央国有资本经营预算基金资助项目(财企[2013]470号)
湖南省自然科学基金资助项目(2017JJ2052)
湖南省科技计划重点基金资助项目(2015JC3053)
教育部产学合作协同育人基金资助项目(201601004010)
湖南省研究生创新基金资助项目(CX2017B112)
关键词
钢轨波磨
弦测法
逆滤波
广域测量
激光位移传感器
rail corrugation
chord-based measuring method
inverse filtering method
wide-area measurement
laser displacement sensor
