基于经验模态分解法的下穿铁路工程对既有轨道变形监测数据降噪方法

Noise Reduction Method of Track Deformation Monitoring Data of Under-crossing Railway Engineering Project Based on EMD

针对下穿铁路工程中既有轨道变形的现场监测数据存在多种因素干扰,难以真实反映下穿铁路工程对既有轨道结构实际影响的问题,将现场监测数据的干扰因素分为测量仪器误差,以及造成既有轨道结构变形的环境温度。根据测量误差、造成既有轨道结构变形的温度,以及下穿铁路工程引起的既有轨道结构附加变形等3类数据组成成分在频域分布集中区域不同的特点,利用EMD(经验模态分解)法分离上述3类既有轨道结构变形监测数据组成成分,剔除既有轨道结构变形监测数据中的干扰因素信号,实现下穿铁路工程中既有轨道变形监测数据的降噪。以某地铁盾构隧道下穿铁路工程现场实测数据为例进行信号分类分析。结果表明,该方法可有效剔除现场变形监测数据中存在的干扰因素成分,从而获得由下穿铁路工程引起的既有轨道结构真实变形。

In view of the fact that there are many interference factors in the actual monitoring data of track deformation of under-crossing railway engineering,it is difficult to reflect the actual influence of under-crossing railway engineering on track structure.The interference factors of actual monitoring data are divided into measurement instrument error and track structure deformation caused by environmental temperature.According to the different characteristics of three types of monitoring data components such as measurement error,temperature of track structure deformation and additional deformation of track structure caused by under-crossing railway engineering in frequency domain,EMD(empirical mode decomposition)is used to se-parate the components of the above three,to eliminate the interference factor signals in track structure deformation monitoring data,and to realize noise reduction of track deformation monitoring data of under-crossing railway engineering.Taking the actual monitoring data of certain metro shield tunnel under-crossing existing railway engineering project as an example,the signal separation results show that this method can effectively eliminate the interference factors in the site monitoring data signal,so that the real deformation of the track structure caused by under-crossing railway project is obtained.