Prediction Method of Track Irregularity Based on EMD

Usually the railway track irregularity is considered as the main vibration source of the wheel rail system, and the primary mission of railway track maintenance and repair work is decrease and eliminate the track irregularity. Accurately predicting the development trend of railway track irregularity can improve the maintenance efficiency of the existing railway line and ensure the train operation safety. According to the random factors and the wavelength of the track irregularity on the traffic state, using empirical mode decomposition (EMD) combined with the method of state probability, the different time sequence of track irregularity are analyzed and validated, obtained several different frequency intrinsic mode function by EMD decomposition method (intrinsic mode function, IMF) and the one remaining trend, then Hilbert transform is performed on each IMF, calculate the instantaneous frequency, to match the different times of the IMF; by irregularity over processing method for each IMF, calculate the state at different time of the track irregularity distribution; the accuracy of two examples show that the method presented in this paper, and analysis the error of prediction result. The results show that the prediction method has good prediction accuracy and can realize the accurate prediction of the track irregularity state of different wave length.