小波变换在高速铁路轨道精调中的应用  

Application of Wavelet Transform in Fine Adjustment of High-speed Railway

作　　者：赵龙[1] 卢建康[1] 刘竹均 

机构地区：[1]中铁二院工程集团有限责任公司,成都610031

出　　处：《铁道工程学报》2017年第4期30-35,共6页Journal of Railway Engineering Society

摘　　要：研究目的:在高速铁路无砟轨道精调内业处理过程中,通常将数据分段并利用人工方法试探性地设定调整量,通过对比、判断和筛选得到最终的调整数据。这种方法不仅需要投入大量的人力和时间,而且数据的质量也因人而异,无法保证数据在整体上的协调一致。而小波变换不仅可以保证数据在整体上的相似性,又可以使得局部数据平滑缓和。本文利用小波变换的方法将不平顺的数据滤掉,保留平滑的数据,并尽量使轨道平顺性的各项指标满足要求,从而实现数据的半自动调整。研究结论:(1)小波变换可以有效地降低高速铁路轨道精调的TQI指数;(2)经过多次小波变换后,所有纵向评价因子的合格率会增大,而横向评价因子的合格率会先收敛到某个值后再发散;(3)经过多次小波变换,数据的调整百分比会增加,但大部分数据的调整量会约束到-1~1 mm之间,为后续的调整留足空间,并且可以防止极限扣件的出现;(4)本算法可成功应用于无砟轨道精调内业处理过程中,自动计算出轨道调整量,并且使得调整错误率低于7%,大大减少了人工干预。Research purposes： Typically, during the indoor- work process of fine adjustment on ballastless track of high- speed railway, data would be partly deposited, adjustments would be set tentatively with manual method, and final result would be gotten by comparison, judgement and screening. However, this method is resource intensive and time consuming. What＇ s more, too much manual intervention makes the quality of data varies, and the overall coordination of data may not be guaranteed. Wavelet transform performs its advantages, not only in ensuring the similarity between global and regional data, but also in smoothing the local data. The objective of the current study in this paper is to use the wavelet transform method to filter out the irregular data, remain the smooth part and ensure the available indicators of smoothness meeting the requirements. By performing this method, the indoor -work would turn into a kind of semi - automatic data processing. Research conclusions： （ 1 ） Wavelet transform could efficiently reduce TQI in fine adjustment of high - speed railway. （2） After many wavelet transforms, the qualification rate of longitudinal evaluation factor will increase, while the qualification rate of horizontal evaluation factor will diverge after a convergenceona specific value. （ 3 ） After many wavelet transforms, the percentage of adjusted data will increase, but the adjustment amount of most data will becontrolled within -1 - -1 ram, which leaves enough room for subsequent adjustment step and prevents that there is no fastener left to be used. （4） The algorithms have been successfully applied in indoor - work process of fine adjustments on ballastless track, and wavelet transform could compute the adjustments automatically, make the error rate of adjustment reduced to less than 7 %, and largely reduce the degree of manual intervention.

关 键 词：小波变换 轨道精调 TQI 

分 类 号：U238[交通运输工程—道路与铁道工程] U213.4