Switch Fault Prediction Based on EMD Sample Entropy-HSMM Model

Abstract

In view of the problems of under-maintenance and over-maintenance in the current railway turnout system, a degradation model and remaining life prediction model based on the HSMM model of EMD sample entropy were proposed. Combined with field experience, taking the power curve as the research object, firstly, the paper divides the corresponding degradation state of each fault with reference to the expert opinion. Then the empirical mode decomposition (EMD) is used to decompose the signal into several intrinsic mode functions (IMF) for data dimension reduction and noise reduction, and the sample entropy of the IMF is calculated to measure its characteristic value. Finally, the HSMM model is established to evaluate the state of the device and the failure theory is used to predict the remaining life. The simulation results demonstrate the effectiveness of the HSMM model based on EMD sample entropy in state recognition and life prediction.