Physical-clone-based prognostics of mechanical structures

Due to the inherent uncertainties caused by the manufacturing process variation, future loading conditions and damage modeling, the lifetime of the mechanical structures or system used in filed condition substantially different from the result in laboratories. The physical clone, newly proposed term in this paper, is a slightly weak structure designed to produce failure earlier than target structures. So it undergoes identical loading as target structures do including non-measurable or unexpected physical phenomena. The time to failure of the physical clone can be tuned by devised method to secure proper prognostic distance for maintenance or golden-time to evacuation. In this paper, the physical-clone-based approach has been developed and demonstrated using 316L stainless steel under high cycle fatigue and compared with model-based approach.

[1]  Julie Bannantine,et al.  Fundamentals of metal fatigue analysis , 1989 .

[2]  M.G. Pecht,et al.  Prognostics and health management of electronics , 2008, IEEE Transactions on Components and Packaging Technologies.

[3]  C. Bailey,et al.  A Physics-of-failure based Prognostic Method for Power Modules , 2008, 2008 10th Electronics Packaging Technology Conference.

[4]  D. Goodman,et al.  A board-level prognostic monitor for MOSFET TDDB , 2006, 2006 IEEE Aerospace Conference.

[5]  Michael Osterman,et al.  In Situ Interconnect Failure Prediction Using Canaries , 2014, IEEE Transactions on Device and Materials Reliability.

[6]  Kai Goebel,et al.  Uncertainty in Steady-State Diagnostics of a Current-Pressure Transducer: How Confident are We in Diagnosing Faults , 2014 .