Development of an Anomaly Detection and Fault Diagnosis Methodology Based on Fault Probability Utilizing Normal Vibration Signals

Purpose: To establish a condition-based maintenance plus (CBM+) and prognostics and health management (PHM) methodology using normal vibration signals in the development of weapon systems.<br/>Methods: A fault probability-based health index using physical features associated with shaft and bearing failure to perform anomaly detection and fault diagnosis.<br/>Results: The methodology was verified using shaft crack test data and bearing acceleration life test data. In shaft crack test data, the shaft health index increased more than the bearing health index. By contrast, in bearing acceleration life test data, the bearing health index increased more than the shaft health index.<br/>Conclusion: This study on designing a normal vibration signal-based anomaly detection and fault diagnosis methodology is expected to broaden the application of CBM+ and PHM in weapon systems.