The Integrity Evaluation of weld zone in railway rails Using Neural Network

신경회로망을 이용한 철도레일 용접부의 건전성평가

  • 윤인식 (경기공업대학 메카트로닉스과) ;
  • 임미섭 (경기공업대학 메카트로닉스과)
  • Published : 2003.06.01


This study proposes the neural network simulator for the integrity evaluation of weld zone in railway rails. For these purposes, the ultrasonic signals for defects(crack) of weld zone in frames are acquired in the type of time series data and echo strength. The detection of the natural defects in railway truck is performed using the characteristics of echodynamic pattern in ultrasonic signal. And then their applications evaluated feature extraction based on the time-frequency-attractor domain(peak to peak, rise time, rise slope, fall time, fall slope, pulse duration, power spectrum, and bandwidth) and attractor characteristics (fractal dimension and attractor quadrant) etc. The constructed neural network simulator agrees fairly well with the measured results of test block(defect location, beam propagation distance, echo strength, etc). The Proposed neural network simulator in this study can be used for the integrity evaluation of weld zone in railway rails.


  1. 윤인식, '초공간을 고려 한 SA508 강의재질열화 시계열 신호의 카오스성 평가,' 대한용접학회지, 제16권, 제6호 pp.537-547, 1998
  2. Thomas, F. P., 'Principles of Radiographic Film Interpretation of Pipeline Welds,' Materials Evaluation, Vol.50, No.11, pp.1268 -1273, 1992
  3. 윤인식, 권성태, '어트랙터 해석을 이용한 레일 용접부의 결함평가,' 한국철도학회지, 제2권, 제1호, pp.38-46, 1999
  4. 윤인식, 권성태, '초음파 파면해석에 의한 대차 프레임의 건전성 평가,' 한국철도학회, 제2권, 제3호, pp.77-83, 2000
  5. 윤인식, 권성태, '위상공간-주파수해석을 이용한 레일 용접부의 결함 평가,' 한국철도학회지, 제2권 제2호, pp.21-30, 1999
  6. 윤인식, 권성태, 박덕신, ;초음파 파면해석에 의한 대차 프레임의 건전성평가,' 한국철도학회 2000년 추계학술대회 논문집, pp.461-467, 2000
  7. Yun In-Sik, 'The Defect Detection and Nondestructive Evaluation in Weld Zone of Austenitic Stainless Steel 304 using Neural Network Ultrasonic Waves,'KSME International Journal, Vo1.12, No.6, pp.295 -306, 1998