A new method to derive rail roughness from axle-box vibration accounting for track stiffness variations and wheel-to-wheel coupling. (1st June 2023)

• Record Type: Journal Article
• Authors: Carrigan, Tobias D.; Talbot, James P.
• Is Part Of: Mechanical systems and signal processing. Volume 192(2023)
• Publication Date: 2023-06-01
• Subjects: Axle-box acceleration -- Acoustic monitoring -- Railway maintenance -- Rail roughness -- Rail corrugation -- Track stiffness; ABA axle-box acceleration -- PSD power spectral density -- ATMS Automated Track Monitoring System -- CAT Corrugation Analysis Trolley; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2023.110232
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: A new method to derive rail roughness spectra, independent of vehicle speed. Model-based transfer function based on stiffness estimates from the P2 resonance. Improved estimates by accounting for multiple wheels on the same rail. Application of a comb filter to remove the effect of wheel roughness. Evaluation via numerical simulation and measurement trials. Abstract: Railways require frequent inspection to maintain acceptable levels of rail roughness. Accelerometers mounted on in-service railway vehicles potentially offer a cost-effective method of near-continuous roughness monitoring by using a transfer function in the frequency domain to derive rail roughness spectra from measurements of axle-box vibration. This paper addresses two phenomena that currently limit the effectiveness of such a method: the variation in track support stiffness along the track, which is the least known property of the vehicle-track system, resulting in a variable transfer function; and the vibration coupling between wheels, which limits the effectiveness of a transfer function based on a single wheel. A new method is presented that estimates the track stiffness by curve-fitting a transfer function to the peak in the axle-box acceleration spectrum associated with the so-called 'P2' resonance of the wheel-rail system. Wheel-to-wheel coupling is addressed by refining the transfer function to account for multiple wheels running on the same rail. Axle-box acceleration is also affected by wheel roughness, and the use of a comb filter is briefly demonstrated to mitigate this effect. The developments are evaluated by both numerical simulation and measurement trials conducted on London Underground's Victoria line.