Gear Fault Diagnosis Based on EMD and AR Spectrum Analysis

A novel method to fault diagnosis of gear crack based on empirical mode decomposition (EMD) and autoregressive (AR) spectrum is presented. This method can carry out empirical mode decomposition and extract feature information of different machine parts in condition monitoring and fault diagnosis of machinery. The main objective of empirical mode decomposition is to separate the time series data into components with different time scale. Then the AR model estimation is applied to each intrinsic mode function and the AR spectrum is obtained. As an example, the vibration signal of a gearbox is analyzed. The experimental results show that this method based on empirical mode decomposition and autoregressive spectrum can effectively diagnose the crack faults of gear.