This paper deals with the localization of cracks in rotating machinery
, based on measured vibrations. The method used is a time domain ident
ification algorithm: the Extended Kalman Filter (EKF). The localizatio
n is performed by designing a bank of EKFs, in which each filter is tu
ned to a different damage hypothesis: i.e., in this case the specific
crack location. By calculating the probabilities of the different hypo
theses, the crack can be localized and its depth can be determined. Th
e procedure is applied to a simulated rotor and to a rotor test rig. (
C) 1996 Academic Press Limited