Hole detection on aluminum plates using inductive learning

摘要

This work discusses the effects of inherent variabilities on the damage identification problem and the creation of a practical damage identification method. Variability is present any time there are factors which have the potential to change during the course of the damage identification process. There are many variabilities which are inherent in damage identification and can cause problems when attempting to detect damage. Manufacturing variability is one of these variabilities and is shown experimentally to be a `non-qualifiable' one. Inductive learning is a tool which has been proposed to be an effective method of performing damage identification. This method is modified to accommodate manufacturing variability and shown to successfully detect hole damage on aluminum plates.