Journal
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
Volume 19, Issue 6, Pages 1666-1684Publisher
SAGE PUBLICATIONS LTD
DOI: 10.1177/1475921719893887
Keywords
Structural health monitoring of rail track; guided wave ultrasound; defect detection; continuously welded rail track; singular value decomposition and independent component analysis
Funding
- CSIR, South Africa
Ask authors/readers for more resources
An experimental monitoring system was installed on an operational heavy haul rail track. The system used two piezoelectric transducers mounted under the head of the rail to transmit and receive ultrasonic guided waves in pulse-echo mode and data were captured over a 2-week period. An artificial defect was introduced by glueing a small mass under the head of the rail at a distance of 370 m from the transducers. The size of the signal reflected by the mass varied as the glue joint deteriorated. The measurements were reordered to simulate a monotonically growing defect. The pre-processing of the captured time signals included averaging, filtering, phased array processing, dispersion compensation, signal stretching and amplitude scaling. Singular value decomposition and independent component analysis of the data were performed. Independent component analysis, with dimension reduction achieved by retaining only the larger principal components, produced the best defect detection. The defect signature was separated as an independent component, and the weight of this component increased monotonically. The results indicate that a transverse defect in the rail head could be detected and located at long range by a system comprising only two transducers. The variation of the signals due to changing environmental and operational conditions limits the size of defect that can be detected, but it is expected that even a relatively small defect, which is significantly smaller than the critical size, would be detected.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available