Characteristic Analysis Of Security State Assessment Data Of High-Speed Train Based On Wavelet Analysis

With the speed of high-speed train continues to improve, the security state of train in the process of long-term service will change and make the train safety warning and health maintenance poses special challenges. During the high-speed train running the sensors monitor a large number of vibration data, how to identify the train running status based on tracking and monitoring data and make security state assessment are critical. Vibration monitoring data is nonlinear and non-stationary signals, wavelet analysis has good time-frequency localization advantages. It is widely used in the signal characteristics analysis. Therefore, in this thesis, analyzing characteristic of security state assessment data of high-speed train based on wavelet analysis, and studying the four states about normal state of train, air spring loss of gas, anti-hunting demolition and lateral damper demolition.In this thesis, firstly, analyzing vibration characteristics of high-speed train’s running gear and the force of spring damping system, based on monitoring data from the time domain, frequency domain and the root-mean-square value, kurtosis statistical parameters for different states signals were analyzed, the three fault states and normal state of vibration differences were diagnosed firstly. The results show that train vibration is low-frequency vibration.Based on continuous wavelet transform grayscale, a feature extraction method of first-order wavelet gray moment was proposed. This method divided different region along wavelet scale and in different region first-order gray moment was calculated. The four kinds of operating states have different first-order wavelet gray moment distribution. In order to analyze the energy spatial distribution of signal, based on wavelet packet transform, this paper improved wavelet packet energy method, the wavelet packet energy moment parameters of vibration signal were extracted. The wavelet packet energy moment of different frequency band was changed and could monitor the train running status. Experimental results show that four states have different energy moment distribution in different speed and different sensor output signal. Finally, the train states were recognized by using support vector machines, and results show that the two kinds of feature extraction algorithms are effective.For the lateral displacement signal has low frequency, combining the wavelet packet to empirical mode decomposition method to extract the characteristics of each state. After the signal was decomposed by wavelet packet, then the last layer of first several frequency band signal was reconstructed, The largest energy band signal of wavelet packet decomposition was decomposed by EMD, and the effective IMF components were reserved based on correlation coefficient, then IMF components were composed to matrix, using singular value decomposition to analyze the matrix, finally completing fault identification based on support vector machines.High-speed train’s state parameter gradients and multiple faults were analyzed based on the first-order wavelet gray moment and wavelet packet energy moment. The distribution characteristic of state process from normal gradient to three kinds of entirely fault was analyzed and also discussed the multiple faults’characteristics were embodied or retained the characteristic of single fault.This work was supported by National Natural Science Foundation of China.(No.61134002)...