Font Size: a A A

Forced response of mistuned bladed disks: Identification and prediction

Posted on:1995-02-10Degree:Ph.DType:Dissertation
University:Arizona State UniversityCandidate:Lin, Chung-ChihFull Text:PDF
GTID:1472390014990551Subject:Engineering
Abstract/Summary:PDF Full Text Request
The present investigation focuses on the dynamic analysis of randomly mistuned bladed disks. In particular, various techniques are employed to obtain reliable estimates of the probability density function of the amplitude of blade response corresponding to a harmonic excitation. First, the applicability of the combined Closed Form-Perturbation (CFP) method is demonstrated for both stiffness and damping mistuning under resonant excitation. To this end, a new numerical algorithm is presented that permits a computationally efficient evaluation of the linearized probability density function of the amplitude of blade vibration. The results of the damping mistuning analysis demonstrate in particular the potentially damaging effects of blade-to-blade changes in damping coefficients and emphasize the need to model such variations.; Next, an adaptive perturbation scheme is introduced for the determination of the forced vibration response of mistuned bladed disks. The versatility of the proposed approach not only guarantees the reliability of the computed response but also leads to an excellent compromise between accuracy and computational effort. The limiting case of a disk supporting an infinite number of blades is considered next to yield the corresponding limit distribution of the amplitude of blade response. This exact asymptotic result can be used in connection with a small size Monte Carlo simulation to produce extremely accurate estimates of the probability density function of blade amplitude in a computationally efficient way.; Finally, the estimation of the parameters of a structural model to represent "at best" a set of measurements of the steady state response of a mistuned bladed disk is achieved by relying on a new technique that combines the advantages of the least squares and maximum likelihood identification methods.
Keywords/Search Tags:Mistuned bladed disks, Response, Probability density function
PDF Full Text Request
Related items