Non-stationary signal analysis and transient machining process condition monitoring

Abstract Non-stationary machine condition monitoring is very important in modern automated manufacturing processes. In this research, an innovative non-stationary (transient) signal analysis approach has been developed for non-stationary machine condition monitoring. It is based on time–frequency distribution analysis and a singular value decomposition approach. The singular value decomposition method is used to extract features from the time–frequency distribution data. These features will serve as machine condition indices and can be easily incorporated for on-line machine condition monitoring and diagnosis. Satisfactory results have been obtained through simulation and experimental data. Experimental studies have demonstrated the effectiveness of the proposed method for transient machine and process condition monitoring.