One of the fundamental problems in data-driven Structural Health Monitoring (SHM) is associated with sourcing data characteristic of the various damage conditions of interest. Where the structure of interest is of high value, as is typically the case in the monitoring of civil and aerospace structures, it is inconceivable that data could be obtained by damaging the real structure. Among the alternatives available is the possibility of generating data from an appropriate physical model, e.g. a Finite Element (FE) model. One of the possible problems here is that the required model fidelity may need to be so high, that model development and benchmarking/updating costs might themselves be prohibitive. In this paper an approach for evaluating the impact of changes in model fidelity is developed and used to investigate the impact of two fidelity factors on feature generation for an experimental structure. The study is based on a model aircraft wingbox component incorporating realistic features: ribs, stringers, rivets and bolts. The two fidelity factors investigated were the mesh density of one of the components, and the impact of FE updating.
An investigation into the necessary model fidelity for SHM feature selection
2008
10 Seiten, 4 Quellen
Aufsatz (Konferenz)
Englisch
FFS Motion Flight Handling Fidelity: A Necessary Step Forward
British Library Conference Proceedings | 2012
|The head rest - A necessary safety-feature
Kraftfahrwesen | 1985
|MID FIDELITY SIMULATION: EXPANDING EXPERIENCE INTO EXPERTISE
TIBKAT | 2019
|A Proposed Model Fidelity Scale.
AIAA | 2008
|