Predictive Health Monitoring of Gear Surface Fatigue Failure Using Model-Based Parametric Method Algorithms; An Experimental Validation

Gears are one of the most important parts of any mechanical transmission system, and in order to achieve reliable operation effective monitoring techniques must be employed. Predictive health monitoring (PHM) systems are currently gaining in popularity due to their effectiveness in providing robust information about the system condition and reducing maintenance costs. However, PHM systems require reliable monitoring techniques, such as vibration, acoustic emission, and oil debris analysis. These techniques have been studied in recent years to discover which can best support the operation of PHM systems in tracing the condition of the operating transmission. These studies have shown the need to apply intelligent algorithms in order to benefit from the advantage of each technique in classifying faults and predicting the onset of failure. This paper presents a new online PHM system for monitoring different gear faults using vibration analysis and autoregressive (AR) algorithms. The intelligent health monitoring system (IHMS) has been implemented on a back-to-back gearbox and can be adapted to monitor the behaviour of transmission systems in automotive, aircraft, wind turbine, and industrial machinery. The study describes the operation of the online IHMS under variable conditions and its capability in detecting transmission gear defects and thus preventing sudden unexpected failure. The results of the experimental test prove the system's capability and support the recent trend of using IHMSs in PHM strategies.