To effectively study vibration characteristics of tracks under different track structures, wavelet transforms of the vibration data are used for pattern classification of vibration feature. First, acceleration data of the track are collected with running speed of 150[Formula: see text]km/h at 26 positions respectively on a slab tangent track, ballast tangent track and ballast curve track by a wireless sensor network (WSN). Then they are analyzed using the power spectral densities (PSDs) and wavelet-based energy spectrum analysis. The paper elaborates on the reasons for the differences of vibration energy and excitation frequencies due to the mechanism of different frequency bands and the corresponding track structures. Based on these, the instantaneous frequencies, vibration energies and durations in the low, medium, and high frequency bands are selected as the features for three track structures. A function curve representing the features is proposed to detect the abnormal track structure by a correlation analysis. Finally, the proposed method of pattern classification has been validated by experimental testings.
Love Thy Neighbour: Automatic Animal Behavioural Classification of Acceleration Data Using the K-Nearest Neighbour Algorithm
