On the comparison of possibilistic and probabilistic methods for the determination of the worst-case response of mistuned bladed disks

Abstract

In this paper, the possibilistic and probabilistic methods are compared with each other by utilizing a lumped parameter vibratory system and a mistuned bladed disk model. In these models stiffness elements are considered to be uncertain, which are assumed to have Gaussian probability distribution. Both methods are utilized for determining uncertain natural frequencies and the worst-case blade tip displacements. The blade tip displacement amplitudes gathered from different number of Monte-Carlo iterations and calculated by fuzzy forced response solution are compared and it is found that among the two, only the possibilistic method converges to the worst-possible blade tip displacement value, and requires significantly less computational time.