Abstract
The paper addresses the issue of online diagnosis and prognosis of emerging faults in human-engineered complex systems. Specifically, the paper reports a dynamic data-driven analytical tool for early detection of incipient faults and real-time estimation of remaining useful fatigue life in polycrystalline alloys. The algorithms for fatigue life estimation rely on time series data analysis of ultrasonic signals and are built upon the principles of symbolic dynamics, information theory and statistical pattern recognition. The proposed method is experimentally validated by using 7075-T6 aluminium alloy specimens on a special-purpose fatigue test apparatus that is equipped with ultrasonic flaw detectors and an optical travelling microscope. The real-time information, derived by the proposed method, is useful for mitigation of widespread fatigue damage and is potentially applicable to life extending and resilient control of mechanical structures.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
