The Application Of Chaos Theory In The Fault Diagnosis Of High-Speed Train

It is an important measure for high speed train to use vibration sensor to monitor the running state of the key parts of high speed train.High speed train monitoring signal is nonlinear and non-stationary. Firstly, we get the phase space reconstruction of the signal,the two key parameters of phase space reconstruction.The optimum delay time and the best embedding dimension is calculated by the false nearest neighbors method and mutual information method. After we have reconstructed phase space,we calculate and analysis two chaotic characteristics (the correlation dimension and the largest lyapunov exponent) and find the train has different chaotic characteristics in4different state value.High speed train fault includes three kinds of typical fault:air spring loss of gas,anti-hunting demolition and lateral damper demolition.We calculate the fault parameter gradient in simulation data.Through the analysis of the simulation data, the experimental results show that the anti-hunting demolition and lateral damper demolition characteristics is obvious. It is proved that the chaos theory for high speed train fault diagnosis is effective.The results show that chaotic analysis method in equipment condition monitoring and fault diagnosis,especially in the fault diagnosis of nonlinear systems shows its unique advantage,which has a broad prospect of application.