In order to further realize the tim

In order to further realize the timely diagnosis and and fault diagnosis exclusion of the performance of high-speed train steering rack special components, and also the improvement of safety and comfort, a dual tree complex wavelet based fault characteristic analysis for high-speed trains is proposed. The characteristic of the signal is extracted based on dual tree complex wavelet to get he characteristic of wave information and multi -scale tendency of fault signals in the process of high-speed railway, and it uses the Hilbert transform of the wavelet coefficients to obtain the wavelet coefficients at each time interval, so as to obtain the self calibration and high computational efficiency of fault characteristic signals in the process of high-speed train steering. Finally, the experiments show that the proposed algorithm could get synchronous lifting of fault recognition rate with the increase of speed, when the speed is greater than 195 km/h, we can get above 90 % of the fault recognition performance, which verifies the effectiveness of the proposed algorithm. Key words: Dual tree complex wavelet; feature extraction; fault diagnosis; bogie 1