The Application Of Chaos Feature To Analysis Of Monitoring Data Of High-speed Train Bogie

High-speed railway is developing rapidly in our country, a series of achievements in technology are achieved, train’s speed was improved constantly. At the same time, more strict techniques are required to train and the safe and stable operation risk will increase accordingly. Train is a continuous running and complex technology system, some key components of the train, especially the performance and fault of the bogie key components can cause serious security hidden danger and accident on the train. To monitor signal of the train body and the bogie, train running state as well as the key components performance are estimated from these vibration signals, vibration signal is of great significance to the safe and stable operation of the train and has great value in engineering application.Vibration For high-speed train running gear, starting from the bogie, the effect factors to train vibration and train vibration transmission is analyzed, this paper introduces the function and working principle of bogie key components (lateral damper, anti-hunting, empty spring) the function and working principle. And through time and frequency domain analysis of different number of lateral damper faults and normal vibration signals of train, so as to have a preliminary understanding of train detection data.The paper is completed based on the the National Natural Science Foundation Project ’Rearch on the security state assessments of high-speed train using monitored data’. The following study is completed:1. Three chaos characteristics (the correlation dimension, Lyapunov index and Kolmogorov entropy) meaning and applicability in bogie fault diagnosis are researched. Using three kinds of chaotic characteristics to constitute high-dimensional feature vector with different number of lateral damper failure, a single anti-hunting failure, parameter gradient condition of bogie key components, then these feature vector of fault types are analyzed and summarized,the experimental results show the effectiveness of the three kinds of chaotic characteristics.2. Paper combines the chaos characteristics and EEMD decomposition method, there are a number of IMF components after EEMD decomposition, keep effective components of the IMF based on the relationship numerical value. Then calculating chaotic characteristics of IMF components, includes the correlation dimension, Kolmogorov entropy and Lyapunov index. Based on this method, the simulation experimental analyse the conditions of train bogie different lateral damper failure and a single lateral damper fault and a single anti-hunting fault at different position etc. Experimentson classify the different conditions with chaotic characteristics by support vector machine (SVM), the experimental results show the effectiveness of this method.3. In view of the high-speed train key parts parameter gradient condition, combining the chaos characteristics(correlation dimension, Kolmogorov entropy and Lyapunov index) and EEMD decomposition method, the simulation experimental analyse the parameter gradient conditions of ateral damper, anti-hunting and empty spring. Compare the method with wavelet chaotic characteristics and IMF entropys, the simulation results show the effectiveness of this method, there are some certain theory reference value on the safety state evaluation of train operation.The research was supported by the Key Project of National Natural Science Foundation of China——Key problem research on high-speed train safty assessment based on data. (No.61134002)...