Speaker: Prof. Jan D. Achenbach

McCormick School of Engineering and Applied Science

Northwestern University

Abstract:

     Damage tolerance under cyclic loading, i.e. the control of time-dependent degradation by prognosis, combined with inspection and timely repair, is an important issue for fatigue damage of safety-critical structures such as aircraft, nuclear reactors and bridges.

      Inspections are conventionally carried out at prescribed intervals by taking non-destructive testing equipment to the structure. Conceptually structural health monitoring (SHM), based on the use of permanently installed sensors that provide information on the state of the structure, continuously or at discrete intervals, is a more effective and economic technique for timely detection of structural deterioration.  Among the issues that will be discussed are the development from NDI to NDE to QNDE to SHM, what is SHM, what is the state of the art, why SHM, and what has to be done to make progress.

     The processing of data from an SHM system necessarily should be done within a probabilistic framework, since little information is available or can be developed reliably by deterministic considerations..

     As an example the growth of a surface-breaking crack under cyclic loading is considered.  The initial depth of the crack is defined statistically by a log-normal distribution.  The growth of the crack is assumed to be defined by Paris law.  Detection of the crack depends on a probability of detection (POD) rule.  The final result of the analysis is the probability, as a function of the number of fatigue cycles, that a crack larger than a critical value is not detected.