铁路轮轨信号内蕴分量的分离新算法  

A New Algorithm for Separation of Intrinsic Components of Railway Wheel-rail Signals

作　　者：李芙蓉 张少刚 马小姝 Li Furong;Zhang Shaogang;Ma Xiaoshu(School of Electronic Information and Electrical Engineering,Tianshui formal University,Tianshui Gansu 741001,China)

机构地区：[1]天水师范学院电子信息与电气工程学院,甘肃天水741001

出　　处：《天水师范学院学报》2020年第5期43-45,共3页Journal of Tianshui Normal University

基　　金：甘肃省高等学校创新能力提升项目“大数据波动内蕴分量规律与研究应用”(2019B-129)阶段性成果。

摘　　要：铁路轮轨信号由多种振动冲击信号复合而成,为分离不同振动冲击信号,可以采用将复合信号转换为复平面曲线的方法。在算法上利用密切曲率圆的拟合替代计算曲率的微分算法,减少噪声和误差对计算结果的影响。该算法将铁路轮轨信号经Hilbert变换,构建出其在复平面上绕动点逆时针转动的平面曲线,利用曲线曲率形成过程对随机振动信号进行分离,对分离得到的每个分量给出其随机时变幅值、随机相位随时间变化的关系,进而为铁路轮轨振动冲击信号应用于工程实践提供一种可借鉴的方法。Railway wheel and rail signals are composed of multiple vibration and shock signals.To separate different vibration and shock signals,a method of converting the composite signal into a complex plane curve can be used.In the algorithm,the close curvature circle fitting is used to replace the differential algorithm for calculating the curvature to reduce the influence of noise and errors on the calculation results.The algorithm was simulated in MATLAB software to verify the feasibility of the algorithm.At the same time,further analysis of the relationship between the amplitude,phase and zero-baseline drift of the random vibration component of each decomposition quantity was performed for the analysis of railway wheel-rail signals and fault A new method is provided for engineering applications such as diagnosis.

关 键 词：铁路轮轨冲击振动信号 微分几何 曲率 零基线漂移 

分 类 号：O324[理学—一般力学与力学基础]