Characteristic Analysis Of Monitoring Data Of High-Speed Train Based On EEMD

With the rapid development of China high-speed train technology, the speed of the train vehicles are constantly improved, the train components wear out more rapidly resulting in the rapid disintegration of its performance parameters, seriously affected the quality of train running,in the process of long-term service. Theose 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,Ensemble Empirical Mode Decomposition (EEMD) with adaptive, local orthogonal, completeness and a series of advantages, has been widely applied in the analysis of signal characteristics. Therefore, in this thesis,analyzing characteristic of security state assessment data of high-speed train based on EEMD and researching the four states about train normal, empty spring loss of air pressure, anti-hunting demolition and lateral damper demolition.(1) This article first analyzes vibration characteristics of high-speed train’s running gear and the force of spring damping system, based on the monitoring data from time domain and frequency domain analysis of train vibration signals of different state, a preliminary diagnosis of the vibration of the three kinds of fault state and normal state difference, get a lot of train vibration is low frequency vibration.(2) Further study of the EEMD and give the definition and physical meaning of singular entropy, energy entropy, average entropy and distance entropy based on the characteristics of the IMF’s time. According to the characteristics of low frequency vibration signal, the EEMD method to explore the characteristics of the various operating states. The simulation data and experimental data for this experiment, after the vibration signal was decomposed by EEMD, keep effective component of the IMF according to the correlation coefficient and extraction entropy as defined IMF entropy, then IMF entropy were composed to vector. Finally, combined with the improved Hyper-sphere support vector machines, in order to realize the high-speed train fault vibration signal classification. Train status identification verify the effectiveness of the five kinds of feature extraction algorithm, on the basis of the high-speed train vibration signal characteristics, analyzed the entropy application in high-speed train fault detection and classification mechanism.(3) Based on the entropy and cross correlation sample entropy characteristics parameter gradient condition are simulation and analyzed, based on the IMF entropy, the simulation analysis of the characteristics of fault conditions, the train from normal gradient to three complete failure characteristics in the process of distribution, and discusses the fault characteristics whether or retain the characteristics of a single faultThis work was supported by National Natural Science Foundation of China.(No.61134002)...