Analysis method for the Measured Track Geometry Data using Wavelet Transform

웨이브렛 변환을 이용한 궤도틀림 분석

  • 이인규 (한국철도기술연구원 궤도토목연구본부) ;
  • 김성일 (한국철도기술연구원 궤도토목연구본부) ;
  • 여인호 (한국철도기술연구원 궤도토목연구본부)
  • Published : 2006.04.01


The regularity of railway track alignment is a crucial component fur maintaining travel safety and the smoothness of passenger ride. The conventional spectral analysis has been considered to estimate the severity of the track irregularity from measured data. The time domain data used to be changed into the frequency domain by Fourier transform. Because the measuring points can be regarded as the time points, the spatial-frequency can be introduced instead of the time-frequency. Although FFT(Fast Fourier Transform) and/or PSD(Power Spectral Density) function could provide fairly localized information within frequency domain, but chronical configurations of data could be missed. In this study, we attempt to apply the Morlet wavelet transform for the purpose of a frequency-time-domain analysis rather than a frequency-domain analysis. The applicability of wavelet transform is examined for the estimation of the track irregularity with real measured track data on the section of Kyoung-bu line by EM-120 measuring vehicle. It is shown that the wavelet transform can be an effective tool to manage the track irregularity.


  1. 알기쉬운 웨이브렛 변환, 이승훈, 윤동한, 진한도서, 2002
  2. 구조물 평가를 위한 신호처리기술, 스마트사회기반시설 연구센터, 2005
  3. Wavelet 해석, 최병선, 세경사, 2001
  4. Esveld, C, 'Modern Railway Track', MRT-Productions, 2001
  5. 시설물의 안전에 관한 유지관리 업무편람, 철도청시설국, 1997
  6. 궤도장비와 선로관리, 서사범, 얼과 알, 2000
  7. 강기동, '고속철도 장파장 궤도틀림 분석에 관한 연구', 한국철도학회 논문집, 제8권 2호, pp.111-115, 2005