Wheel-flat diagnostic tool via wavelet transform

The detection and acknowledgement of signatures, for condition monitoring and fault diagnostics by wavelet transform, deserves increased attention, due to its property of variable time–frequency resolution, which overcomes limitations of classical time–frequency approaches.In the paper, a diagnostic tool is presented, based on the wavelet transform, able to detect and to quantify the wheel-flat defect of a test train at different speeds and to measure the train speed with proper accuracy. The designed diagnostic tool minimises the hardware requirements, since only one accelerometer is needed, and provides results in real time.The results, achieved by an exhaustive experimental campaign, permit to validate the effectiveness of the diagnostic tool and to demonstrate the advantages of wavelet-based detection of signatures.