Zili LiRolf Dollevoet
Noise and Vibration
Mechanical Systems and Signal Processing
Volume 64, Pages 224-242
Publishing date: August 27, 2014
baseline railway track, field hammer test measurements, frequency response based statistical method, health monitoring of railway tracks, squats' characteristic frequencies
Link or Download
In this paper, the application of the Frequency Response Function (FRF) based statistical method is proposed to identify the characteristic frequencies of damaged conditions in railway tracks. The method compares the damaged state to a healthy state based on non-destructive field hammer test measurements. First, a study is carried out to investigate the repeatability of hammer tests in railway tracks. By changing the excitation and measurement locations it is shown that the variability introduced by the test process is negligible. Second, following the concepts of control charts employed in process monitoring, a method to define an approximate healthy state is introduced by using hammer test measurements at locations without visual damage. Then, a case study of squats (i.e.\ a mayor type of rail surface defects) of different severity is carried out. The identified frequency ranges related to squats agree with those found in an extensively validated vehicle-borne detection system. Therefore, the FRF-based statistical method in combination with the non-destructive hammer test measurements can be employed to identify the characteristic frequencies of damaged conditions in railway tracks in the frequency range of 300-3000 Hz. These frequencies are useful for the development of vehicle-borne monitoring systems based on dynamic response measurements.