A guided wave based monitoring system for welded freight rail, has previously been developed. The existing system was designed to only detect complete breaks. Current research efforts are focused on including a pulse-echo mode of operation in order to detect, locate and possibly monitor damage before a complete break occurs. For this application, it would be advantageous to distinguish between cracks, which could result in rail breaks, and other scattering defects which do not represent damage such as welds. We investigate the interaction between long-range propagating modes and selected weld and crack geometries in an attempt to relate scattering behaviour to defect geometry. Four candidate modes, suitable for long range propagation, are identified and evaluated. A weighted average reflection coefficient is defined and used as a measure to quantify mode conversion between these four modes. The results show that it should be possible to distinguish between a large crack in the crown of the rail and a weld. We also present a procedure to convert reflection and transmission modal amplitudes, computed in the frequency domain, to time domain signals.
Reference:
Long, C.S and Loveday, P.W. 2014. Numerical evaluation of pulse-echo damage detection in train rail. In: 9th South African Conference on Computational and Applied Mechanics, Somerset West, 14–16 January 2014
Long, C. S., & Loveday, P. W. (2014). Numerical evaluation of pulse-echo damage detection in train rail. http://hdl.handle.net/10204/7547
Long, Craig S, and Philip W Loveday. "Numerical evaluation of pulse-echo damage detection in train rail." (2014): http://hdl.handle.net/10204/7547
Long CS, Loveday PW, Numerical evaluation of pulse-echo damage detection in train rail; 2014. http://hdl.handle.net/10204/7547 .